diff --git a/.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.i18n.yaml b/.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.i18n.yaml
new file mode 100644
index 0000000000..ef424471d8
--- /dev/null
+++ b/.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.i18n.yaml
@@ -0,0 +1,6 @@
+# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
+# side as of the last confirmed-consistent state. Both languages carry equal authority;
+# after editing either side, bring the other along and re-record with:
+#   pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md
+2026-08-10-fork-children-stay-one-shot.md: 030bed3c1b516a54afd5f97b00000ed7ceaf64d8
+2026-08-10-fork-children-stay-one-shot.zh.md: 4b033a70c3674315c491e85be67a1d357aeac51b
diff --git a/.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md b/.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md
new file mode 100644
index 0000000000..030bed3c1b
--- /dev/null
+++ b/.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md
@@ -0,0 +1,49 @@
+# Agent Note: Forked children stay one-shot
+
+Status: implemented
+
+English | [中文](2026-08-10-fork-children-stay-one-shot.zh.md)
+
+## Problem
+
+Fork's only difference from spawn is that the child Session is seeded with the parent's completed-turn prefix ([subagent-fork](../../../../packages/subagent/subagent-fork/README.md)). That seed costs real tokens — the inherited history is re-sent in every child request — and its one concrete payoff is provider-side prefix reuse: under the same provider and model, a child request whose leading bytes are identical to the parent's re-prefills none of the shared span. Anything a child scope adds *ahead* of the inherited history spends that payoff, because reuse stops at the first differing byte.
+
+The child-scoped `report` return channel is now the largest such addition, and since [the report obligation](../feature/2026-08-06-continuable-child-report-obligation.md) it is two deltas rather than one: the `report` tool schema and the `tool:report` system-prompt section. Both live in the request head — the system block and the tool block precede every message — so a continuable forked child invalidates reuse before the first inherited turn and re-prefills the whole transcript it was forked to reuse. That composition pays fork's duplication cost and collects none of its benefit, while the parent still holds a reusable prefix the child could have shared.
+
+## Decision
+
+Every shipped composition binds the fork delegation tool to `backgroundMode: one-shot`: [the base bundle](../../../../packages/bundle/base/cordis.patch.yml), [the ACP example](../../../../examples/acp-agent/cordis.yml), and [the headless example](../../../../examples/headless-agent/cordis.yml). The base bundle leaves `run_in_background` available, because it mounts a task service; the two examples set `enableRunInBackground: false`, because they mount none and a one-shot background start would otherwise fail at call time on a missing `tasks` service.
+
+One-shot children — foreground and background alike — are created through `SubagentService.start()`, which never enters the continuable activation-setup registry, so neither `report` nor its prompt section is installed. A forked one-shot child's system prompt and tool schemas therefore equal its parent's, apart from the `persona` and `toolFilter` deltas a deployment opts into per delegation tool.
+
+`spawn` keeps `backgroundMode: continuable`. Continuable children and the report obligation ship unchanged for the provider whose child starts with no inherited prefix to protect, so this decision costs the report channel nothing.
+
+### The restriction is composition, not code
+
+`ForkProvider.prepareContinuable` stays implemented and `ctx.subagents.startContinuable()` still accepts `fork`; only the shipped `cordis.yml` rows changed. `tool-subagent` knows both the provider's `inheritsParentContext` and its own `backgroundMode` at mount, so a load-time rejection of the pair was available and is deliberately not added: the pair is not wrong in general. It is wrong only while a child-scope delta precedes inherited history, and the package that creates that delta — [`dsh-tool-subagent-report`](../../../../packages/subagent/tool-subagent-report/README.md) — is separately installable and, by its own design, invisible to `tool-subagent`. A deployment that omits the report package can run continuable forked children with the prefix intact. Encoding one roster's consequence as a delegation-tool invariant would make the tool assert something it cannot observe.
+
+The reintroduction condition is recorded as a `TODO(fork-continuable-prefix-reuse)` marker on `prepareContinuable` itself, the one method the shipped compositions do not call, and tracked as issue #2124: continuable fork reopens when a child's system prompt and tool schemas can match its parent's byte for byte.
+
+## Alternatives considered
+
+**Reject `inheritsParentContext` + `continuable` at mount.** A loud load-time failure would prevent silent reintroduction, which is what the configuration change cannot do. Rejected because the delegation tool cannot see the report package and the combination is legitimate without it; the invariant would be false for a deployment that never installs a child-scope delta, and `tool-subagent` would be asserting a fact owned by the roster.
+
+**Stop mounting the fork provider at all.** This was the broader form of the restriction. Rejected because foreground fork *is* the prefix-reusing case and is untouched by the report channel, so a full ban gives up the capability without buying anything the one-shot binding does not already buy — and would leave no shipped composition exercising session seeding.
+
+**Ship continuable forked children and accept the loss.** Rejected because the loss is total rather than marginal: reuse breaks ahead of the inherited history, so the child pays full prefill on a transcript it duplicated for the sole purpose of not paying it. A deployment that wants a long-lived child with no inherited context already has `spawn`.
+
+**Make `report` visible to every Agent.** A global registration would restore byte-identical prefixes by giving parent and child the same schema and section. Rejected because roots, one-shot children, remote children, and agentless callers would advertise a tool with no derivable recipient, and execution-time rejection would make schema visibility disagree with authority — the scope-local decision the [report tool Agent Note](../feature/2026-07-30-continuable-subagent-report-tool.md) already settled.
+
+**Install the child-scope deltas after the inherited history.** Rejected as unrepresentable: the system prompt and the tool schemas are request-head structures in every provider's wire format, so no ordering within them can place a child-only addition behind the message list.
+
+## Consequences
+
+- No shipped composition creates a continuable forked child; `subagent_fork` returns a result to its caller's turn, and `send_message` addresses only spawned children.
+- A forked child's request prefix stays byte-identical to its parent's unless the deployment configures `persona` or `toolFilter` on the fork delegation tool, so the token cost of seeding buys provider-side reuse again.
+- The fork provider's continuable path has no production caller and no assembled-composition coverage. It keeps its package-level tests, and the seam still accepts it, so a bundle or `--patch` overlay can reintroduce it with no code change and no warning.
+- `subagent_fork`'s model-visible schema changes: the continuable background wording is replaced by the one-shot task wording in the base bundle, and disappears entirely from the two examples. The affected keyless snapshot tool-schema sidecars are re-recorded in the same change.
+- The report obligation's reach narrows to spawned children in shipped deployments. Its default `wakeup` scheduling, authority model, and coverage are unchanged.
+
+### Accepted risks
+
+The constraint lives in three configuration files and a code comment, not in a gate. A future bundle row or profile patch can set `backgroundMode: continuable` on a fork tool and silently reintroduce the prefix loss; nothing fails loud. That is the accepted cost of not encoding one roster's consequence into `tool-subagent`.
diff --git a/.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.zh.md b/.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.zh.md
new file mode 100644
index 0000000000..4b033a70c3
--- /dev/null
+++ b/.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.zh.md
@@ -0,0 +1,49 @@
+# Agent Note: fork 出的 child 保持 one-shot
+
+Status: implemented
+
+[English](2026-08-10-fork-children-stay-one-shot.md) | 中文
+
+## 问题
+
+fork 与 spawn 的唯一区别是 child 的 Session 会以 parent 已完成轮次的前缀作为初始内容（见 [subagent-fork](../../../../packages/subagent/subagent-fork/README.md)）。这份初始内容有实打实的 token 成本——继承的历史会在 child 的每次请求中重新发送——而它唯一确定的回报是提供方侧的前缀复用：在提供方与模型相同的前提下，起始字节与 parent 逐字节相同的 child 请求，无需为这段共享区间重新预填充。任何由 child 作用域添加在继承历史*之前*的内容都会消耗掉这份回报，因为复用在第一个不同字节处即告停止。
+
+作用域局部的 `report` 返回通道现在是此类添加中最大的一项，而自[report 义务](../feature/2026-08-06-continuable-child-report-obligation.md)起它是两项而非一项增量：`report` 工具 schema，以及 `tool:report` 系统提示词 section。两者都位于请求头部——系统块与工具块先于所有消息——因此一个可继续的 fork child 会在第一条继承轮次之前就使复用失效，并重新预填充它当初 fork 就是为了复用的整份 transcript（文本记录）。这种组合付出了 fork 的复制成本却收不到它的收益，而 parent 手上仍握着一份 child 本可共享的可复用前缀。
+
+## 决策
+
+所有随附组合都把 fork 委派工具绑定为 `backgroundMode: one-shot`：[base 组合包](../../../../packages/bundle/base/cordis.patch.yml)、[ACP 示例](../../../../examples/acp-agent/cordis.yml)与[headless 示例](../../../../examples/headless-agent/cordis.yml)。base 组合包保留 `run_in_background`，因为它挂载了 task 服务；两个示例设置 `enableRunInBackground: false`，因为它们都不挂载 task 服务，否则一次 one-shot 后台启动会在调用时因缺少 `tasks` 服务而失败。
+
+one-shot child——前台与后台皆然——经由 `SubagentService.start()` 创建，该路径从不进入可继续的 activation setup 注册表，因此 `report` 与它的提示词 section 都不会被安装。于是一个 fork 出的 one-shot child 的系统提示词与工具 schema 与其 parent 相同，只差部署逐个委派工具主动选择的 `persona` 与 `toolFilter` 增量。
+
+`spawn` 保持 `backgroundMode: continuable`。对于 child 起步时本就没有继承前缀需要保护的那个提供方，可继续 child 与 report 义务随附行为不变，因此本决策没有让 report 通道付出任何代价。
+
+### 该限制在于组合，不在于代码
+
+`ForkProvider.prepareContinuable` 仍然实现完好，`ctx.subagents.startContinuable()` 也仍接受 `fork`；改动的只有随附的 `cordis.yml` 行。`tool-subagent` 在挂载时同时知道提供方的 `inheritsParentContext` 与自身的 `backgroundMode`，因此一个加载期拒绝该组合的检查是可行的，而这里刻意不加：该组合并非普遍错误。它只在某个 child 作用域增量位于继承历史之前时才是错的，而产生该增量的包——[`dsh-tool-subagent-report`](../../../../packages/subagent/tool-subagent-report/README.md)——是独立安装的，并且按其自身设计对 `tool-subagent` 不可见。一个不安装 report 包的部署可以在前缀完好的前提下运行可继续的 fork child。把某一份插件清单的后果写成委派工具的不变量，会让该工具断言它无法观察到的事实。
+
+重新开放的条件记录为 `prepareContinuable` 方法上的 `TODO(fork-continuable-prefix-reuse)` 标记——随附组合不调用这个方法——并由 issue #2124 跟踪：当 child 的系统提示词与工具 schema 能与其 parent 逐字节一致时，可继续 fork 即可重新开放。
+
+## 备选方案
+
+**在挂载时拒绝 `inheritsParentContext` 与 `continuable` 的组合。** 一次响亮的加载期失败可以阻止悄然的重新引入，而配置改动做不到这一点。否决的原因是委派工具看不到 report 包，且在没有它时该组合是合法的；对于从不安装任何 child 作用域增量的部署，这个不变量是假的，而 `tool-subagent` 会去断言一件由插件清单拥有的事实。
+
+**干脆不挂载 fork 提供方。** 这是该限制更彻底的形式。否决的原因是前台 fork *正是*复用前缀的那种情形，且不受 report 通道影响，因此全面禁用会在不换来任何 one-shot 绑定尚未换来的东西的同时放弃该能力——并且随附组合将没有任何一个演练 session 初始内容。
+
+**照常随附可继续的 fork child 并接受这份损失。** 否决的原因是这份损失是全额而非边际的：复用在继承历史之前就已中断，于是 child 为一份自己复制过来、目的恰恰是不必付费的 transcript 付了全额预填充。想要一个没有继承上下文的长期 child 的部署，本来就有 `spawn`。
+
+**让 `report` 对每个 Agent 可见。** 全局注册会通过让 parent 与 child 拥有相同的 schema 与 section 来恢复逐字节相同的前缀。否决的原因是根 agent、one-shot child、远端 child 与无 agent 调用方都会宣告一件推导不出收件方的工具，而执行期拒绝会让 schema 可见性与权限彼此矛盾——这正是[report 工具 Agent Note](../feature/2026-07-30-continuable-subagent-report-tool.md)已经定下的作用域局部决策。
+
+**把 child 作用域增量安装到继承历史之后。** 否决的原因是它无法表达：在每个提供方的协议格式中，系统提示词与工具 schema 都是请求头部结构，因此它们内部的任何排序都无法把仅属于 child 的添加放到消息列表之后。
+
+## 后果
+
+- 没有任何随附组合会创建可继续的 fork child；`subagent_fork` 把结果返回给调用方的轮次，而 `send_message` 只寻址 spawn 出的 child。
+- 除非部署在 fork 委派工具上配置了 `persona` 或 `toolFilter`，fork child 的请求前缀与其 parent 逐字节相同，因此初始内容的 token 成本重新换来了提供方侧的复用。
+- fork 提供方的可继续路径没有生产调用方，也没有整体组装层面的覆盖。它保留自己的包内测试，seam 也仍然接受它，因此某个组合包或 `--patch` 覆盖层可以无需改动代码、也不会有任何警告地把它重新引入。
+- `subagent_fork` 面向模型的 schema 发生变化：base 组合包中可继续的后台措辞被 one-shot 的 task 措辞取代，在两个示例中则完全消失。受影响的无密钥快照工具 schema 伴随文件在同一次改动中重新记录。
+- 在随附部署中，report 义务的覆盖范围收窄到 spawn 出的 child。它的 `wakeup` 默认调度、权限模型与覆盖均保持不变。
+
+### 已接受的风险
+
+该限制存在于三个配置文件与一处代码注释中，而不在门禁里。未来某个组合包行或 profile 补丁可以在 fork 工具上设置 `backgroundMode: continuable`，从而悄然重新引入前缀损失；没有任何东西会失败得很响亮。这就是不把某一份插件清单的后果写入 `tool-subagent` 所接受的代价。
diff --git a/.agents/notes/implemented/bug-fix/2026-08-06-list-agents-residency-vocabulary.i18n.yaml b/.agents/notes/implemented/bug-fix/2026-08-06-list-agents-residency-vocabulary.i18n.yaml
new file mode 100644
index 0000000000..5f01146dbe
--- /dev/null
+++ b/.agents/notes/implemented/bug-fix/2026-08-06-list-agents-residency-vocabulary.i18n.yaml
@@ -0,0 +1,6 @@
+# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
+# side as of the last confirmed-consistent state. Both languages carry equal authority;
+# after editing either side, bring the other along and re-record with:
+#   pnpm run verify-translation-pairing --write .agents/notes/implemented/bug-fix/2026-08-06-list-agents-residency-vocabulary.md
+2026-08-06-list-agents-residency-vocabulary.md: 01fff958921465909f5cd300dc355fdac6ec0772
+2026-08-06-list-agents-residency-vocabulary.zh.md: d68b9857de23ccb0b68f0bfafba153fde63303a7
diff --git a/.agents/notes/implemented/bug-fix/2026-08-06-list-agents-residency-vocabulary.md b/.agents/notes/implemented/bug-fix/2026-08-06-list-agents-residency-vocabulary.md
new file mode 100644
index 0000000000..01fff95892
--- /dev/null
+++ b/.agents/notes/implemented/bug-fix/2026-08-06-list-agents-residency-vocabulary.md
@@ -0,0 +1,40 @@
+# Agent Note: `list_agents` uses `ready` for resumable children
+
+Status: implemented
+
+English | [中文](2026-08-06-list-agents-residency-vocabulary.zh.md)
+
+## Problem
+
+`list_agents` projected a continuable child's process residency as `running | idle | complete`. `complete` reads as a terminal unit of work with a result somewhere, but the underlying fact says only that no Activation is resident: the conversation is intact, `send_message` can continue it, and nothing about the child's outcome is being claimed. A model that reads `complete` reasonably looks for a result to collect or sends replacement work to a conversation it believes has ended.
+
+The word is especially misleading alongside [manager-owned settlement delivery](../feature/2026-08-06-manager-owned-subagent-settlement-delivery.md). Completion reaches the parent as a notice; listing exists to recall durable conversations, not to poll for that notice.
+
+## Decision
+
+The model-facing projection reports `running | idle | ready`:
+
+- **`running`** means the resident Agent has an active driver.
+- **`idle`** means the Agent is resident between turns and may be waiting on agents it started.
+- **`ready`** means only the durable conversation remains. `send_message` starts the next turn on the same conversation; the status is resumable rather than terminal and does not mean a result is waiting to be collected.
+
+The tool description states those distinctions and directs the model away from polling: it says the parent is told when a child finishes and that listing is for recalling which children it started. `send_message` remains the authoritative delivery check because either snapshot may race another process or a later message.
+
+The service layer is unchanged. `SubagentListEntry.activity` retains `'running' | 'inactive'`, which accurately describes corpus residency for consumers such as a UI. The model-facing adapter maps `inactive` to `ready` because that word communicates the action available to the model without inventing an outcome.
+
+## Alternatives considered
+
+**Keep `complete` and qualify it in the description.** A description saying that `complete` does not mean complete fights the rendered status on every read. The line the model scans must carry the correct distinction itself.
+
+**Use `active | dormant`.** This removes the useful distinction between a resident Agent that is between turns and a storage-only conversation, and makes the storage-only state sound unavailable. `ready` states the useful fact: the same conversation accepts another turn.
+
+**Drop the status entirely.** Residency remains useful when a parent decides whether to send more work. Removing it trades one misleading status for no signal.
+
+**Rename the service activity values.** `running | inactive` is correct at the service layer and has non-model consumers. Renaming it would churn a general contract to fix one adapter's presentation; the [durable catalog note](../feature/2026-07-22-durable-subagent-catalog-and-list-agents.md) continues to own that service vocabulary.
+
+## Consequences
+
+- The rendered line uses `<id> [running] — <label>`, `<id> [idle] — <label>`, or `<id> [ready] — <label>`.
+- The output schema's `status` enum changes with the rendered contract. The generated tool catalog picks up the new description; it renders each tool's `parameters` only and never carried the output schema.
+- Unit coverage pins all three mappings and the description clauses that direct the model to the settlement notice instead of polling this tool.
+- The assembled ACP `subagent-list-agents` scenario renders `ready` for a settled, resumable child.
diff --git a/.agents/notes/implemented/bug-fix/2026-08-06-list-agents-residency-vocabulary.zh.md b/.agents/notes/implemented/bug-fix/2026-08-06-list-agents-residency-vocabulary.zh.md
new file mode 100644
index 0000000000..d68b9857de
--- /dev/null
+++ b/.agents/notes/implemented/bug-fix/2026-08-06-list-agents-residency-vocabulary.zh.md
@@ -0,0 +1,40 @@
+# Agent Note: `list_agents` uses `ready` for resumable children
+
+Status: implemented
+
+[English](2026-08-06-list-agents-residency-vocabulary.md) | 中文
+
+## Problem
+
+`list_agents` 把可继续 child 的进程驻留状态投影为 `running | idle | complete`。`complete` 读起来像一项终态工作，且结果就在某处，但底层事实只表示没有驻留的 Activation：对话完好无损，`send_message` 可以继续它，而且它对 child 的结果不作任何断言。读到 `complete` 的模型会合理地寻找可收集结果，或向一个它以为已经结束的对话发送替代工作。
+
+这个词与[由管理器负责的结算投递](../feature/2026-08-06-manager-owned-subagent-settlement-delivery.md)同时出现时尤其容易误导。完成会以通知到达 parent；列表用于回忆持久化对话，而不是轮询该通知。
+
+## Decision
+
+面向模型的投影报告 `running | idle | ready`：
+
+- **`running`** 表示驻留 Agent 存在活跃 driver。
+- **`idle`** 表示 Agent 驻留但处于轮次之间，也可能正在等待它启动的 agent。
+- **`ready`** 表示只剩下持久化对话。`send_message` 会在同一对话上启动下一轮；该状态表示可恢复而非终态，也不表示有结果等待收集。
+
+工具描述会陈述这些区别，并引导模型远离轮询：它说明 child 结束时 parent 会收到通知，而列表用于回忆自己启动过哪些 child。由于任一快照都可能与另一进程或后续消息竞态，`send_message` 仍是投递时的权威检查。
+
+服务层不变。`SubagentListEntry.activity` 保留 `'running' | 'inactive'`，对 UI 等消费方而言，这准确描述了语料驻留状态。面向模型的适配器把 `inactive` 映射为 `ready`，因为这个词表达了模型可执行的操作，而没有虚构结果。
+
+## Alternatives considered
+
+**保留 `complete`，并在描述中限定它。** 一段解释 `complete` 不代表完成的描述，每次被读取时都在与渲染状态对抗。模型扫读的那一行必须自身表达正确区别。
+
+**使用 `active | dormant`。** 这会删除处于轮次之间的驻留 Agent 与仅存于存储的对话之间的有效区别，并让仅存于存储的状态听起来不可用。`ready` 直接表达有用事实：同一对话可以接受下一轮。
+
+**完全移除状态。** parent 决定是否发送更多工作时，驻留状态依然有用。移除它只是用没有信号替代一个误导性状态。
+
+**重命名服务活动值。** `running | inactive` 在服务层是正确的，并且有非模型消费方。为了修复一个适配器的呈现而搅动通用契约并不合理；[持久化目录 Agent Note](../feature/2026-07-22-durable-subagent-catalog-and-list-agents.md) 继续拥有该服务词汇。
+
+## Consequences
+
+- 渲染行使用 `<id> [running] — <label>`、`<id> [idle] — <label>` 或 `<id> [ready] — <label>`。
+- 输出 schema 的 `status` 枚举与渲染契约一同变化。生成的工具目录会带上新描述；它只渲染每个工具的 `parameters`，从来不收录输出 schema。
+- 单元覆盖固定三种映射，以及引导模型等待结算通知而非轮询本工具的描述条款。
+- 整体组装的 ACP `subagent-list-agents` 场景会为已结算且可恢复的 child 渲染 `ready`。
diff --git a/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.i18n.yaml b/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.i18n.yaml
index 75e87605ec..d6440dd126 100644
--- a/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.i18n.yaml
@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write .agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.md
-2026-07-21-log-backed-session-titles.md: 47025ba7e72ea6746d6f4428a81d6d754a8dcd03
-2026-07-21-log-backed-session-titles.zh.md: d7e12875354ca1fc66f0035020c7ffe2ddaa9354
+2026-07-21-log-backed-session-titles.md: 46128a3e53d6617c9fa7a0c27d1ac01acfd1d9c1
+2026-07-21-log-backed-session-titles.zh.md: 2cdf5c22d6b2a28b27a68472619ced4b7dedd106
diff --git a/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.md b/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.md
index 47025ba7e7..46128a3e53 100644
--- a/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.md
+++ b/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.md
@@ -44,7 +44,7 @@ Automatic provider failures are nonfatal warnings and retain the latest title. E
 
 A fork inherits seed title events unchanged, like the rest of its source log — a pinned (user-sourced) title stays pinned in the child until an explicit refresh. The first-message provider does not automatically retitle a fork. The all-messages provider may append a child-owned revision after a later child prompt, using inherited and new eligible messages.
 
-`ctx.sessionQuery.readTitle()` folds one live-preferred or persisted log without loading titles during `listSessions()`. The TUI uses the latest title as its header subtitle and sets the terminal window title to `<session title> — <configured product title>` after terminal-safe rendering. The Web host folds the same log state into a validated mux control frame after each attached-session subscription baseline and immediately after forwarding a live raw title event. The browser retains only newer title event seqs even when the control frame precedes list or session-instance creation; sidebar labels, search, breadcrumbs, and the browser title then react to the projected revision. `session.list` remains metadata-only, so a cold persisted session uses the cwd basename or id until opening or resuming it attaches the log. The browser title uses `<session title> — <existing HTML title>` only for a selected titled session and otherwise preserves the product title. Consumers reporting agent completion use the core `findLastMessageTurnEnd()` fold, so a later between-turn title record cannot replace the preceding message-triggered outcome.
+`ctx.sessionQuery.readTitle()` folds one live-preferred or persisted log without loading titles during `listSessions()`. The TUI uses the latest title as its header subtitle and sets the terminal window title to `<session title> — <configured product title>` after terminal-safe rendering. The Web host folds the same log state into a validated mux control frame after each attached-session subscription baseline and immediately after forwarding a live raw title event. The browser retains only newer title event seqs even when the control frame precedes list or session-instance creation; sidebar labels, search, breadcrumbs, and the browser title then react to the projected revision. `session.list` remains metadata-only, so a cold persisted session uses the cwd basename or id until opening or resuming it attaches the log. The browser title uses `<session title> — <existing HTML title>` only for a selected titled session and otherwise preserves the product title. Consumers reporting agent completion use the core `foldConsumedWork()` fold, so a later between-turn title record cannot replace the preceding message-triggered outcome.
 
 ## Alternatives considered
 
diff --git a/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.zh.md b/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.zh.md
index d7e1287535..2cdf5c22d6 100644
--- a/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.zh.md
+++ b/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.zh.md
@@ -44,7 +44,7 @@ Status: implemented
 
 与源日志的其他部分相同，fork 会原样继承作为种子的标题事件——被钉住（user 来源）的标题在子会话中保持钉住，直到显式 refresh。首消息提供方不会自动为 fork 重新生成标题。全部消息提供方可以在子会话出现后续提示词后追加一项归子会话所有的修订，并使用继承的合格消息和新增的合格消息。
 
-`ctx.sessionQuery.readTitle()` 会折叠一份实时优先或已持久化的日志，而不会在 `listSessions()` 期间加载标题。TUI 使用最新标题作为其标题栏副标题，并在完成终端安全渲染后，将终端窗口标题设置为 `<session title> — <configured product title>`。Web host 会在每个已附加会话的订阅基线之后，以及转发实时原始标题事件后立即，将同一份日志状态折叠为经过校验的 mux 控制帧。即使控制帧先于列表或会话实例创建抵达，浏览器也只保留标题事件 seq 较新的版本；侧边栏标签、搜索、面包屑和浏览器标题会随投影后的修订更新。`session.list` 仍只包含元数据，因此尚未打开的持久化会话会继续以 cwd 基名或 id 作为回退，直至打开或恢复会话时附加其日志。浏览器仅在选中已有标题的会话时将标题设置为 `<session title> — <existing HTML title>`，否则保留产品标题。报告 agent 完成情况的消费方使用核心的 `findLastMessageTurnEnd()` 折叠逻辑，因此后续的轮次间标题记录无法取代此前由消息触发的结果。
+`ctx.sessionQuery.readTitle()` 会折叠一份实时优先或已持久化的日志，而不会在 `listSessions()` 期间加载标题。TUI 使用最新标题作为其标题栏副标题，并在完成终端安全渲染后，将终端窗口标题设置为 `<session title> — <configured product title>`。Web host 会在每个已附加会话的订阅基线之后，以及转发实时原始标题事件后立即，将同一份日志状态折叠为经过校验的 mux 控制帧。即使控制帧先于列表或会话实例创建抵达，浏览器也只保留标题事件 seq 较新的版本；侧边栏标签、搜索、面包屑和浏览器标题会随投影后的修订更新。`session.list` 仍只包含元数据，因此尚未打开的持久化会话会继续以 cwd 基名或 id 作为回退，直至打开或恢复会话时附加其日志。浏览器仅在选中已有标题的会话时将标题设置为 `<session title> — <existing HTML title>`，否则保留产品标题。报告 agent 完成情况的消费方使用核心的 `foldConsumedWork()` 折叠逻辑，因此后续的轮次间标题记录无法取代此前由消息触发的结果。
 
 ## 考虑过的替代方案
 
diff --git a/.agents/notes/implemented/feature/2026-07-22-durable-subagent-catalog-and-list-agents.i18n.yaml b/.agents/notes/implemented/feature/2026-07-22-durable-subagent-catalog-and-list-agents.i18n.yaml
index 538e89e047..e97d76c979 100644
--- a/.agents/notes/implemented/feature/2026-07-22-durable-subagent-catalog-and-list-agents.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-22-durable-subagent-catalog-and-list-agents.i18n.yaml
@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write .agents/notes/implemented/feature/2026-07-22-durable-subagent-catalog-and-list-agents.md
-2026-07-22-durable-subagent-catalog-and-list-agents.md: 61bda2a52f2ea5db0a582fe668643a13b967f091
-2026-07-22-durable-subagent-catalog-and-list-agents.zh.md: 6b5bff468349df4226aaf76dd7ed7d9ce5c3f33c
+2026-07-22-durable-subagent-catalog-and-list-agents.md: 51664bc1e211802ef61f47508b9e57fdd099b9e7
+2026-07-22-durable-subagent-catalog-and-list-agents.zh.md: 79503a30e081abffd45637439ac539d95ae794f0
diff --git a/.agents/notes/implemented/feature/2026-07-22-durable-subagent-catalog-and-list-agents.md b/.agents/notes/implemented/feature/2026-07-22-durable-subagent-catalog-and-list-agents.md
index 61bda2a52f..51664bc1e2 100644
--- a/.agents/notes/implemented/feature/2026-07-22-durable-subagent-catalog-and-list-agents.md
+++ b/.agents/notes/implemented/feature/2026-07-22-durable-subagent-catalog-and-list-agents.md
@@ -23,7 +23,7 @@ Parent-to-child enumeration is a service capability with consumer-specific proje
 - report corpus activity separately as `running` or `inactive`, without implying completion or resumability;
 - return every resulting child in stable `createdAt` ascending, child-id ascending order.
 
-Every ordinary local start receives a `one-shot` descriptor with an optional caller-owned display label, while the continuation manager persists a labeled `continuable` descriptor containing its additional reconstruction fields. The model-facing delegation tool already owns a short `description` and supplies it for one-shot display; lower-level callers such as workflows need not invent presentation metadata. The model-facing `list_agents` adapter filters the service result to continuable children and refines status through the live Agent registry (`running`/`idle`/`complete`); a UI can consume both modes and choose an id-based fallback for unlabeled one-shot history. Descriptor persistence, by-id lookup, direct-parent authorization, and provider-independent cold resume remain owned by the implemented Activation contract. Listing consumes those facts but cannot weaken them or invent a second descriptor representation.
+Every ordinary local start receives a `one-shot` descriptor with an optional caller-owned display label, while the continuation manager persists a labeled `continuable` descriptor containing its additional reconstruction fields. The model-facing delegation tool already owns a short `description` and supplies it for one-shot display; lower-level callers such as workflows need not invent presentation metadata. The model-facing `list_agents` adapter filters the service result to continuable children and refines status through the live Agent registry (`running`/`idle`, and [`ready`](../bug-fix/2026-08-06-list-agents-residency-vocabulary.md) for storage-only); a UI can consume both modes and choose an id-based fallback for unlabeled one-shot history. Descriptor persistence, by-id lookup, direct-parent authorization, and provider-independent cold resume remain owned by the implemented Activation contract. Listing consumes those facts but cannot weaken them or invent a second descriptor representation.
 
 ### Enumeration decision
 
@@ -52,7 +52,7 @@ If measured scale later requires an index, that index is derived state: session
 
 A valid descriptor produces one child entry, a per-child inspection failure produces one diagnostic entry, and a candidate without a descriptor produces no entry. `mode` is durable creation policy; `activity` is a process-local corpus snapshot. Activity is neither `AgentStatus`, the manager's internal Activation state, nor a durable outcome, and the result does not expose the internal `createdAt` sorting key. Exact Activation states and durable outcomes such as successful completion, failure, cancellation, and stop reason require a separate durable activation record and are outside this feature.
 
-The model-facing `list_agents` tool takes one optional `scope: 'children' | 'descendants'` argument, derives the root id from the current execution Agent, and resolves the request through an explicit request-to-spec step (`undefined` → `children`) before either execution or rendering. The resolved `children` scope calls `SubagentService.listChildren(rootSessionId)`, while `descendants` calls `SubagentService.listDescendants(rootSessionId)`. Its internal output projection keeps `id` and `parent` as branded `SessionId` values until the tool JSON boundary. It keeps diagnostics, drops `one-shot` child entries, derives status from the live Agent registry — `running` for an active driver, `idle` for a resident Agent between turns, and `complete` when no live Agent remains — then renders `<id> [<status>] — <label>` or `<id> [diagnostic: <reason>]` in stable catalog order. The `descendants` scope flattens the complete tree from one live-preferred corpus in stable pre-order, traverses ordinary and one-shot intermediates so deeper continuable agents are discovered, revalidates each cold candidate against its enumerated lifecycle, and adds `parentId`/`depth` to every entry. The tool inserts ` parent=<id> depth=<n>` before the label; `parent` is the durable direct-parent session id and may name an omitted ordinary session. For the current caller, only depth-1 child rows are `send_message` candidates, while deeper child rows may be selected for `interrupt_agent` ([interrupt contract](2026-08-06-continuable-subagent-interrupt.md)). Discovery is a hint only — follow-up authority stays exact-direct-parent, and interrupt authority stays with the service's live-lineage check. An empty projection renders `(no subagents)`.
+The model-facing `list_agents` tool takes one optional `scope: 'children' | 'descendants'` argument, derives the root id from the current execution Agent, and resolves the request through an explicit request-to-spec step (`undefined` → `children`) before either execution or rendering. The resolved `children` scope calls `SubagentService.listChildren(rootSessionId)`, while `descendants` calls `SubagentService.listDescendants(rootSessionId)`. Its internal output projection keeps `id` and `parent` as branded `SessionId` values until the tool JSON boundary. It keeps diagnostics, drops `one-shot` child entries, derives status from the live Agent registry — `running` for an active driver, `idle` for a resident Agent between turns, and `ready` when no live Agent remains — resumable rather than terminal — then renders `<id> [<status>] — <label>` or `<id> [diagnostic: <reason>]` in stable catalog order. The `descendants` scope flattens the complete tree from one live-preferred corpus in stable pre-order, traverses ordinary and one-shot intermediates so deeper continuable agents are discovered, revalidates each cold candidate against its enumerated lifecycle, and adds `parentId`/`depth` to every entry. The tool inserts ` parent=<id> depth=<n>` before the label; `parent` is the durable direct-parent session id and may name an omitted ordinary session. For the current caller, only depth-1 child rows are `send_message` candidates, while deeper child rows may be selected for `interrupt_agent` ([interrupt contract](2026-08-06-continuable-subagent-interrupt.md)). Discovery is a hint only — follow-up authority stays exact-direct-parent, and interrupt authority stays with the service's live-lineage check. An empty projection renders `(no subagents)`.
 
 In the superseded trace-based path, diagnostics used three fixed reasons. Malformed event surfaces, conflicting headers discovered during an exact child load, a read result whose immutable header differs from the traced candidate or no longer names the requested direct parent, a target that is no longer the located descriptor event, malformed descriptor content, and multiple descriptor events mapped to `corrupt`. An unknown descriptor version mapped to `unsupported`. `SESSION_QUERY_SESSION_NOT_FOUND`, `SESSION_QUERY_EVENT_NOT_FOUND`, and `SESSION_QUERY_PERSISTENCE_FAILED` from a per-child read mapped to `unavailable`. This phase boundary was intentional: a persistence outage during the initial trace failed the operation, while the same outage beginning during candidate reads could produce one identical `unavailable` diagnostic per affected child; the first version neither coalesced those diagnostics nor promoted them to a global failure. A missing descriptor was instead a non-subagent exclusion without a diagnostic. Configuration/window errors and unrecognized failures were not child diagnostics and propagated as operation failures. Each diagnostic identified the child id and reason without exposing model-hidden descriptor content; the candidate was omitted while healthy siblings remained visible. Sessions outside the trace's direct descendants were never read and produced no diagnostic.
 
@@ -95,7 +95,7 @@ The first version has no child deletion operation. If later product behavior del
 - `packages/subagent/subagent/tests/service.spec.ts` pins descriptor v2 parsing for both modes and proves an unlabeled raw start resolves a one-shot descriptor before provider dispatch. `packages/subagent/subagent-inprocess/tests/subagent-inprocess.spec.ts` proves the local driver appends that descriptor inside the initial turn, returns the published id when cancellation lands in the factory-to-run handoff, and keeps result and handle-disposal failures on separate channels. Delegation-tool tests pin propagation of their existing display description and preserve independent result and disposal diagnostics.
 - `packages/subagent/subagent/tests/list-children.spec.ts` pins the current read path against a real composition of the session store, JSONL persistence, spawn/fork providers, the subagent service, and the projection registry — no query service — keylessly: live-only listing without persistence; loud `SUBAGENT_CONTROL_PROJECTIONS_UNAVAILABLE` and `SUBAGENT_CONTROL_SESSION_STORE_UNAVAILABLE` even with zero children; the three-rung ladder (a live child never inspected, a cold child inspected exactly once, and the cache-hit, absent-key, absent-service, and poisoned-row second-rung cases); last-wins over multiple descriptors; malformed payloads and unknown versions diagnosed as `corrupt`; a failed cold inspection as one `unavailable` diagnostic retried on the next listing; a fork seed's ancestor descriptor listed under that identity; foreign-unit fold failures contained per child as `corrupt` on both the live and cold paths; `createdAt`-then-id ordering without ordinary forks; provider absence without child omission; compacted/uncompacted twins listing identically; a persisted-listing failure failing the whole enumeration; cancellation normalized to stable `CANCELLED`; typed stable error codes; and descendant listing's iterative stable pre-order, traversal through ordinary and one-shot intermediates, positioned diagnostics, lifecycle revalidation, and cancellation. A companion spec (retired together with the query-backed read path) rejected eager evaluation of the optional session-query runtime while importing the ordinary subagent surface.
 - `packages/subagent/tool-subagent-control/tests/list-agents.spec.ts` pins the `list_agents` schema (one optional `scope` enum), the continuable-only projection that omits a healthy one-shot sibling while preserving diagnostics, registry-derived child/diagnostic/empty text forms, an end-to-end settled-child listing with its durable label, the descendants scope's pre-order parent/depth annotations across a live waiting branch, cancellation forwarding to both scopes, the no-agent rejection, the `agents` load requirement without `sessionQuery`, and HMR disposal.
-- The keyless ACP snapshot scenario `subagent-list-agents` (examples/acp-agent) fences its second parent turn on a snapshot-only `subagent/end` marker, then executes `list_agents` for real against the subagent service, the projection registry, and JSONL persistence, rendering `<id> [complete] — <label>`.
+- The keyless ACP snapshot scenario `subagent-list-agents` (examples/acp-agent) fences its second parent turn on a snapshot-only `subagent/end` marker, then executes `list_agents` for real against the subagent service, the projection registry, and JSONL persistence, rendering `<id> [ready] — <label>`.
 - The keyless snapshot scenario `subagent-diagnostic` (examples/headless-agent) pins the current listing's model-visible diagnostic classification, including a descriptor-less settled child surfacing as a `corrupt` diagnostic.
 - The keyless ACP snapshot scenario `subagent-published-run-failure` publishes a real one-shot child, injects independent run-result and handle-disposal failures, and preserves both diagnostics in the parent tool result.
 
diff --git a/.agents/notes/implemented/feature/2026-07-22-durable-subagent-catalog-and-list-agents.zh.md b/.agents/notes/implemented/feature/2026-07-22-durable-subagent-catalog-and-list-agents.zh.md
index 6b5bff4683..79503a30e0 100644
--- a/.agents/notes/implemented/feature/2026-07-22-durable-subagent-catalog-and-list-agents.zh.md
+++ b/.agents/notes/implemented/feature/2026-07-22-durable-subagent-catalog-and-list-agents.zh.md
@@ -23,7 +23,7 @@ parent 到 child 的枚举是一项带消费方专用投影的服务功能。`Su
 - 将语料活动状态单独报告为 `running` 或 `inactive`，但不暗示已完成或可恢复；
 - 按 `createdAt` 升序、再按 child id 升序稳定返回所有结果 child。
 
-每次普通的本地启动都会收到带可选、由调用方拥有之显示标签的 `one-shot` 描述符，而继续执行管理器会持久化带标签、包含附加重建字段的 `continuable` 描述符。面向模型的委派工具已经拥有简短 `description`，会将其用于一次性显示；workflow 等底层调用方无需凭空构造展示元数据。面向模型的 `list_agents` 适配器会将服务结果过滤为可继续 child，并通过在线 Agent 注册表细化状态（`running`／`idle`／`complete`）；UI 可以消费两种模式，并为无标签的一次性历史选择基于 id 的回退展示。描述符持久化、按 id 查找、直接 parent 鉴权和不依赖提供方的冷恢复仍归已实现的 Activation 约定负责。列表查询消费这些事实，但不能削弱它们，也不能另行发明第二种描述符表示。
+每次普通的本地启动都会收到带可选、由调用方拥有之显示标签的 `one-shot` 描述符，而继续执行管理器会持久化带标签、包含附加重建字段的 `continuable` 描述符。面向模型的委派工具已经拥有简短 `description`，会将其用于一次性显示；workflow 等底层调用方无需凭空构造展示元数据。面向模型的 `list_agents` 适配器会将服务结果过滤为可继续 child，并通过在线 Agent 注册表细化状态（`running`／`idle`，以及对应仅存于存储的 [`ready`](../bug-fix/2026-08-06-list-agents-residency-vocabulary.md)）；UI 可以消费两种模式，并为无标签的一次性历史选择基于 id 的回退展示。描述符持久化、按 id 查找、直接 parent 鉴权和不依赖提供方的冷恢复仍归已实现的 Activation 约定负责。列表查询消费这些事实，但不能削弱它们，也不能另行发明第二种描述符表示。
 
 ### 枚举决策
 
@@ -52,7 +52,7 @@ subagent 服务将 `sessionQuery` 保持为可选依赖，因此没有该服务
 
 有效描述符产生一个 child 条目，逐 child 检查失败产生一个 diagnostic 条目，缺少描述符的候选不产生条目。`mode` 是持久化创建策略；`activity` 是进程本地语料快照。活动状态既不是 `AgentStatus`、管理器内部的 Activation 状态，也不是持久化结果，结果不公开内部 `createdAt` 排序键。成功完成、失败、取消和停止原因等精确 Activation 状态与持久化结果需要单独的持久化激活记录，不在本功能范围内。
 
-面向模型的 `list_agents` 工具接受一个可选的 `scope: 'children' | 'descendants'` 参数，从当前执行 Agent 推导根 id，并在执行或渲染前通过显式的 request-to-spec 步骤解析请求（`undefined` → `children`）。解析后的 `children` scope 调用 `SubagentService.listChildren(rootSessionId)`，`descendants` scope 则调用 `SubagentService.listDescendants(rootSessionId)`。其内部输出投影中的 `id` 与 `parent` 会一直保持为品牌化的 `SessionId` 值，直到工具 JSON 边界。它保留 diagnostic，丢弃 `one-shot` child 条目，状态取自在线 Agent 注册表——driver 活跃为 `running`，驻留但处于轮次之间为 `idle`，没有在线 Agent 时为 `complete`——然后按稳定目录顺序渲染 `<id> [<status>] — <label>` 或 `<id> [diagnostic: <reason>]`。`descendants` scope 从一份实时优先语料按稳定 pre-order 展平完整树，遍历普通与一次性中间节点以发现更深的可继续 agent，依据枚举生命周期重新校验每个冷候选，并为每个条目附加 `parentId`／`depth`。工具会在 label 之前插入 ` parent=<id> depth=<n>`；`parent` 是持久化直接 parent 会话 id，可能指向被省略的普通会话。对于当前调用方，只有 depth-1 child 条目可作为 `send_message` 候选，更深的 child 条目则可供 `interrupt_agent` 选择（[中断约定](2026-08-06-continuable-subagent-interrupt.md)）。发现结果只是提示——follow-up 权限仍仅属于确切直接 parent，中断权限仍由服务的在线 lineage 检查决定。空投影渲染为 `(no subagents)`。
+面向模型的 `list_agents` 工具接受一个可选的 `scope: 'children' | 'descendants'` 参数，从当前执行 Agent 推导根 id，并在执行或渲染前通过显式的 request-to-spec 步骤解析请求（`undefined` → `children`）。解析后的 `children` scope 调用 `SubagentService.listChildren(rootSessionId)`，`descendants` scope 则调用 `SubagentService.listDescendants(rootSessionId)`。其内部输出投影中的 `id` 与 `parent` 会一直保持为品牌化的 `SessionId` 值，直到工具 JSON 边界。它保留 diagnostic，丢弃 `one-shot` child 条目，状态取自在线 Agent 注册表——driver 活跃为 `running`，驻留但处于轮次之间为 `idle`，没有在线 Agent 时为 `ready`（可恢复而非终态）——然后按稳定目录顺序渲染 `<id> [<status>] — <label>` 或 `<id> [diagnostic: <reason>]`。`descendants` scope 从一份实时优先语料按稳定 pre-order 展平完整树，遍历普通与一次性中间节点以发现更深的可继续 agent，依据枚举生命周期重新校验每个冷候选，并为每个条目附加 `parentId`／`depth`。工具会在 label 之前插入 ` parent=<id> depth=<n>`；`parent` 是持久化直接 parent 会话 id，可能指向被省略的普通会话。对于当前调用方，只有 depth-1 child 条目可作为 `send_message` 候选，更深的 child 条目则可供 `interrupt_agent` 选择（[中断约定](2026-08-06-continuable-subagent-interrupt.md)）。发现结果只是提示——follow-up 权限仍仅属于确切直接 parent，中断权限仍由服务的在线 lineage 检查决定。空投影渲染为 `(no subagents)`。
 
 在已被取代的追踪读路径中，diagnostic 使用三种固定原因。格式错误的事件 surface、精确加载 child 时发现的 header 冲突、读取结果中的不可变 header 与追踪到的候选不一致或不再指向请求的直接 parent、读取目标不再是先前定位的描述符事件、格式错误的描述符内容和多个描述符事件映射为 `corrupt`。未知描述符版本映射为 `unsupported`。逐 child 读取产生的 `SESSION_QUERY_SESSION_NOT_FOUND`、`SESSION_QUERY_EVENT_NOT_FOUND` 和 `SESSION_QUERY_PERSISTENCE_FAILED` 映射为 `unavailable`。这项阶段边界是有意为之：初始追踪期间发生持久化故障会让操作失败，而同一故障如果始于候选读取期间，可能会让每个受影响的 child 分别产生一条相同的 `unavailable` diagnostic；第一版既不合并这些 diagnostic，也不会把它们提升为全局失败。缺少描述符则作为非 subagent 排除，且不产生 diagnostic。配置错误、窗口错误和未识别的失败不属于 child diagnostic，会作为操作失败继续向上传播。每条 diagnostic 都标识 child id 及原因，不暴露对模型隐藏的描述符内容；系统会排除该候选，而其他健康的 sibling 仍然可见。系统绝不会读取不属于追踪结果直接后代的会话，也不会为它们产生 diagnostic。
 
@@ -95,7 +95,7 @@ diagnostic 是瞬时查询结果，不属于会话事件或目录状态。推导
 - `packages/subagent/subagent/tests/service.spec.ts` 固定两种模式下的描述符 v2 解析，并证明无标签的底层启动会在分发给提供方之前解析出一次性描述符。`packages/subagent/subagent-inprocess/tests/subagent-inprocess.spec.ts` 证明本地驱动会在初始轮次内追加该描述符，在取消落入工厂到 run 的交接窗口时返回已发布 id，并让结果与句柄释放失败保留在独立通道中。委派工具测试固定其现有显示说明的传递，并保留相互独立的结果与 dispose diagnostic。
 - `packages/subagent/subagent/tests/list-children.spec.ts` 针对由会话存储、JSONL 持久化、spawn/fork 提供方、subagent 服务与投影注册表构成的真实组合——不含查询服务——以无密钥方式钉住现行读取路径：无持久化时的仅存活列表；零 children 也响亮报 `SUBAGENT_CONTROL_PROJECTIONS_UNAVAILABLE` 与 `SUBAGENT_CONTROL_SESSION_STORE_UNAVAILABLE`；三级阶梯（存活 child 从不检查、冷 child 恰好检查一次，以及缓存命中、key 缺席、服务缺席、行中毒四个第二级用例）；多描述符 last-wins 取末者；载荷格式错误与未知版本诊断为 `corrupt`；冷检查失败成一条 `unavailable` diagnostic 并在下次列表重试；fork seed 中的祖先描述符按该身份列出；外部 unit 折叠失败在存活与冷两条路径上按 child 收纳为 `corrupt`；按 `createdAt` 再按 id 排序且不列普通 fork；提供方缺失时不排除 child；压缩与未压缩的孪生 child 列表结果一致；持久化列表失败使整次枚举失败；取消稳定归一化为 `CANCELLED`；带类型的稳定错误码；以及后代列表的迭代式稳定 pre-order、穿过普通与一次性中间节点、带位置 diagnostic、生命周期复验与取消。一个伴随规格(已随查询式读取路径一起退役)曾在导入普通 subagent surface 时拒绝对可选 session-query 运行时的 eager 求值。
 - `packages/subagent/tool-subagent-control/tests/list-agents.spec.ts` 固定 `list_agents` 的 schema（一个可选 `scope` 枚举）、只保留可继续 child 且排除健康的一次性 sibling、同时保留 diagnostic 的投影、由注册表推导的 child／diagnostic／空结果文本形式、带持久化 label 的已结束 child 端到端列表、descendants scope 在在线 waiting 分支上的 pre-order parent/depth 注释、两个 scope 的取消信号转发、无调用 agent 时的拒绝、要求 `agents` 但不再注入 `sessionQuery` 的加载约定，以及 HMR dispose。
-- 无密钥 ACP 快照场景 `subagent-list-agents`（examples/acp-agent）使用仅限快照的 `subagent/end` 标记为第二个 parent 轮次设置边界，随后针对 subagent 服务、投影注册表和 JSONL 持久化真实执行 `list_agents`，渲染 `<id> [complete] — <label>`。
+- 无密钥 ACP 快照场景 `subagent-list-agents`（examples/acp-agent）使用仅限快照的 `subagent/end` 标记为第二个 parent 轮次设置边界，随后针对 subagent 服务、投影注册表和 JSONL 持久化真实执行 `list_agents`，渲染 `<id> [ready] — <label>`。
 - 无密钥快照场景 `subagent-diagnostic`（examples/headless-agent）钉住现行列表的模型可见诊断分类，包括无描述符的定局 child 以 `corrupt` diagnostic 出现。
 - 无密钥 ACP 快照场景 `subagent-published-run-failure` 会发布一个真实的一次性 child，注入相互独立的 run result 与 handle dispose 失败，并在 parent 工具结果中保留两项 diagnostic。
 
diff --git a/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.i18n.yaml b/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.i18n.yaml
index 775794a381..d250012eb9 100644
--- a/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.i18n.yaml
@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write .agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.md
-2026-07-30-continuable-subagent-report-tool.md: 6a607f1da48fdfde36b2d9351f131dd3ef67ba7f
-2026-07-30-continuable-subagent-report-tool.zh.md: b4d65fc71695e9fca46f4e2c69c9749745de260a
+2026-07-30-continuable-subagent-report-tool.md: d4681d0ea30335a5537525927ecad615e6edd8b8
+2026-07-30-continuable-subagent-report-tool.zh.md: 731225d2a5b788997704fafe9331901ca9b84393
diff --git a/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.md b/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.md
index 6a607f1da4..d4681d0ea3 100644
--- a/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.md
+++ b/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.md
@@ -12,7 +12,7 @@ Treating every final assistant message as an implicit result would conflate turn
 
 ## Decision
 
-Add the independently installed `@deepseek-ai/dsh-tool-subagent-report` package. It contributes an ordinary model-facing `report` tool to each continuable in-process child Activation. A child may call it zero or multiple times in a turn. Success neither concludes the turn, settles the Activation, nor prevents later parent follow-ups, and finishing a turn never reports automatically.
+Add the independently installed `@deepseek-ai/dsh-tool-subagent-report` package. It contributes an ordinary model-facing `report` tool to each continuable in-process child Activation. The mechanism accepts zero or multiple calls in a turn; the child is separately instructed to call it once before finishing ([the report obligation](2026-08-06-continuable-child-report-obligation.md)). Success neither concludes the turn, settles the Activation, nor prevents later parent follow-ups, and finishing a turn never reports automatically.
 
 The feature is a collaboration control, not a result-bearing execution wrapper. It adds no Task, `SubagentRun`, result promise, Activation state, delivery queue, or replay path.
 
@@ -22,7 +22,7 @@ The feature is a collaboration control, not a result-bearing execution wrapper.
 
 `messageId` is the stable `MessageId` of the user-role message accepted by the parent. It is not an `InboxItemId`: quiet delivery creates no inbox occurrence, while waking delivery creates one occurrence for the same stable message. It is also not a read receipt, parent-log acknowledgement, turn-completion receipt, or persistence flush.
 
-The description states that reporting is explicit, repeatable, direct-parent-only, and non-terminal. It warns that a failed tool result may still follow an accepted send because a later `tools/post-execute` failure can replace the result. Without an idempotency key, stronger wording would encourage duplicate retries after ambiguous failure.
+The description states that reporting is required before finishing, repeatable, direct-parent-only, and non-terminal. It warns that a failed tool result may still follow an accepted send because a later `tools/post-execute` failure can replace the result. Without an idempotency key, stronger wording would encourage duplicate retries after ambiguous failure.
 
 The tool uses generic rendering with no locations. Its acknowledgement includes `messageId`. Scope-local registration keeps presentation and execution aligned: roots, one-shot children, remote providers, sibling scopes, and agentless execution neither see nor execute `report`. It installs after the child's global `toolFilter`, so a delegation allow-list cannot accidentally remove the structural return channel; deployments that require no return channel omit the package.
 
@@ -36,7 +36,7 @@ Nested reporting crosses exactly one edge. A grandchild reports to its direct ch
 
 ### Delivery policy
 
-The package validates `reportDelivery: 'quiet' | 'wakeup'`; the default is `quiet`.
+The package validates `reportDelivery: 'quiet' | 'wakeup'`; the default is `wakeup` ([why the default reversed](2026-08-06-continuable-child-report-obligation.md)).
 
 Quiet delivery calls `parent.inject()`. It adds model-visible context without starting a parent model request: an idle parent appends before the call returns, while an admitting or running parent stages the report for the next safe log position. It creates no inbox occurrence and therefore no synthetic continuation-manager acceptance record.
 
@@ -56,13 +56,13 @@ The subagent seam adds `registerContinuableSetup(contribution): () => void`, bac
 
 The registry owns registration, per-child installation records, setup rollback, child-scope cleanup, and immediate revocation. Applying a batch returns the Agent setup commit that revalidates provisioning after every setup await and immediately before Agent publication. A throwing or concurrently revoked contribution therefore rejects before either Agent or Session publication and rolls back the batch. New registrations affect a resident child only on its next Activation; removing a registration first closes it to new setup and then revokes every provisioning or resident installation immediately. Registration disposal and child-context disposal are idempotent and attempt every release before aggregating failures.
 
-This seam keeps the continuation manager unaware of tool names. The report package installs only `report`; `@deepseek-ai/dsh-tool-subagent-control` independently installs parent-side `send_message` and `list_agents`. A deployment can install either direction, both, or neither. Providers remain data-only, durable descriptors do not snapshot report availability or delivery mode, and cold resume uses the deployment's current contributions and policy.
+This seam keeps the continuation manager unaware of tool names. The report package installs only `report` and its child-scoped guidance section; `@deepseek-ai/dsh-tool-subagent-control` independently installs parent-side `send_message` and `list_agents`. A deployment can install either direction, both, or neither. Providers remain data-only, durable descriptors do not snapshot report availability or delivery mode, and cold resume uses the deployment's current contributions and policy.
 
 ### Snapshot coverage
 
-The ACP snapshot harness adds `waitForSubagentTurnEnd`, selecting the Nth harvested child by the same order as `session.N.jsonl`. It waits for a closed child turn containing a request header so a continuable child's earlier descriptor-seed turn cannot satisfy the boundary. This lets the assembled quiet-mode scenario wait for the child-side report without inventing a parent-visible signal.
+The ACP snapshot harness adds `waitForSubagentTurnEnd`, selecting the Nth harvested child by the same order as `session.N.jsonl`. It waits for a closed child turn containing a request header so a continuable child's earlier descriptor-seed turn cannot satisfy the boundary. This lets the assembled scenario wait for the child-side report without inventing a parent-visible signal.
 
-The authored snapshot starts a continuable child, executes the real scope-local `report` tool, confirms that the idle parent is not woken, and then submits a later parent prompt that consumes the framed report. It declares child schema pin `1`, so the otherwise non-global `report` schema is checked against `tool-schemas.1.expected.json` while the root keeps the default schema pin. The generated tool catalog separately mints a child scope to include the same scope-local schema.
+The authored snapshot starts a continuable child, executes the real scope-local `report` tool, observes the one ordinary parent turn the default waking delivery creates, and then submits a later parent prompt that consumes the framed report. It declares child pins `1`, so the otherwise non-global `report` schema and the child's own prompt are checked against `tool-schemas.1.expected.json` and `system-prompt.1.expected.md` while the root keeps the class pins. The generated tool catalog separately mints a child scope to include the same scope-local schema.
 
 ## Alternatives considered
 
@@ -72,7 +72,7 @@ Automatic delivery cannot represent zero reports, progress reports, or several s
 
 ### Always wake the parent
 
-Waking on every report creates unsolicited turns and can cascade through nested subagents. Quiet delivery matches background coordination better as the default, while deployments that require immediate action can select wakeup.
+Waking on every report creates unsolicited turns and can cascade through nested subagents. Quiet delivery was chosen as the default on the assumption that the parent had another reason to read its context. [The report obligation](2026-08-06-continuable-child-report-obligation.md) supersedes that choice: a parked background coordinator has no such reason, so waking is the default and this paragraph now records why `quiet` still exists.
 
 ### Let the child choose the delivery mode
 
@@ -103,16 +103,16 @@ A post-creation revocation check can reject the Activation only after the Agent
 - A continuable in-process child exposes exactly one scope-local `report` schema only while the report package's contribution is installed; unrelated Agents never expose it.
 - The tool returns the parent message's stable `MessageId`. Quiet delivery has no `InboxItemId`; waking delivery has a separate inbox occurrence.
 - Only the exact resident child may report, and only to the exact live direct parent derived from durable lineage. The service has no recipient parameter or offline fallback.
-- Quiet delivery is the validated default and never starts a parent request. Wakeup creates exactly one later FIFO turn and never steers an open turn.
+- Waking delivery is the validated default: it creates exactly one later FIFO turn and never steers an open turn. Quiet delivery never starts a parent request.
 - Child cancellation or disposal after parent acceptance does not retract the report. Before acceptance, child disposal, drain, parent loss, or caller cancellation rejects the operation.
 - Fresh and resumed Activations compose current setup contributions before publication. Grants wait for the next Activation; revocation is immediate for resident children.
 - Unit coverage pins visibility, allow-list behavior, both delivery modes, stable message and sender identities, nested routing, invalid senders, absent parents, cancellation, drain, revocation races, and the absence of Tasks or implicit final reporting.
-- The keyless assembled snapshot proves the real child tool, quiet non-wakeup behavior, durable parent framing, and later parent consumption.
+- The keyless assembled snapshot proves the real child tool, the one waking parent turn, durable parent framing, and later parent consumption.
 
 ### Accepted risks
 
 The acceptance boundary is weaker than durable end-to-end delivery. A crash can leave the result ambiguous, and retries may duplicate reports.
 
-Wakeup mode can amplify model work when nested children report frequently. Deployment ownership and a quiet default limit but do not remove that risk.
+Waking delivery can amplify model work when nested children report frequently. Deployment ownership through `reportDelivery` bounds but does not remove that risk.
 
 Registry presence is the parent liveness signal. A host-owned parent whose `AgentHandle.dispose()` has started but has not yet unwound its scope can still accept and append a report that it will not act on in this process. Closing that gap requires an Agent-level disposal-start signal rather than subagent-layer inference.
diff --git a/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.zh.md b/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.zh.md
index b4d65fc716..731225d2a5 100644
--- a/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.zh.md
+++ b/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.zh.md
@@ -12,7 +12,7 @@ Status: implemented
 
 ## 决策
 
-新增可独立安装的 `@deepseek-ai/dsh-tool-subagent-report` 包。它会向每个可继续进程内 child Activation 贡献一个普通的面向模型 `report` 工具。child 在一个轮次中可调用零次或多次。调用成功既不会结束该轮次或结算 Activation，也不会阻止 parent 之后继续 follow-up；完成轮次也绝不会自动报告。
+新增可独立安装的 `@deepseek-ai/dsh-tool-subagent-report` 包。它会向每个可继续进程内 child Activation 贡献一个普通的面向模型 `report` 工具。机制本身接受一个轮次中调用零次或多次；child 会另行被要求在结束前调用一次（见[报告义务](2026-08-06-continuable-child-report-obligation.md)）。调用成功既不会结束该轮次或结算 Activation，也不会阻止 parent 之后继续 follow-up；完成轮次也绝不会自动报告。
 
 该功能是协作控制，不是承载结果的执行包装层。它不新增 Task、`SubagentRun`、结果 promise、Activation 状态、投递队列或回放路径。
 
@@ -22,7 +22,7 @@ Status: implemented
 
 `messageId` 是 parent 接受的用户角色消息所对应的稳定 `MessageId`。它不是 `InboxItemId`：静默投递不创建 inbox 条目实例，唤醒投递则会为同一条稳定消息创建一个条目实例。它也不是已读回执、parent 日志确认、轮次完成回执或持久化 flush。
 
-工具描述会明确报告操作是显式、可重复、仅限直接 parent 且不会结束轮次的。它还会警告：发送被接受后，后续 `tools/post-execute` 失败可能替换工具结果，因此工具结果失败时内容仍可能已经送达。没有幂等键时，更强的表述会诱导调用方在结果不明确的失败后重复重试。
+工具描述会明确报告操作在结束前必须执行、可重复、仅限直接 parent 且不会结束轮次。它还会警告：发送被接受后，后续 `tools/post-execute` 失败可能替换工具结果，因此工具结果失败时内容仍可能已经送达。没有幂等键时，更强的表述会诱导调用方在结果不明确的失败后重复重试。
 
 该工具使用不带 location 的通用渲染，其确认中包含 `messageId`。作用域局部注册使呈现与执行保持一致：root、one-shot child、远程提供方、同级作用域和无 agent（智能体）执行既不能看到，也不能执行 `report`。它会在 child 的全局 `toolFilter` 之后安装，因此委派 allow-list 不会意外移除这条结构性返回通道；不需要返回通道的部署不安装该包。
 
@@ -36,7 +36,7 @@ root、one-shot child、伪造对象、陈旧 Agent 和同 id 替换对象都以
 
 ### 投递策略
 
-该包会校验 `reportDelivery: 'quiet' | 'wakeup'`，默认值为 `quiet`。
+该包会校验 `reportDelivery: 'quiet' | 'wakeup'`，默认值为 `wakeup`（见[默认值反转的理由](2026-08-06-continuable-child-report-obligation.md)）。
 
 静默投递调用 `parent.inject()`。它会添加模型可见上下文，但不启动 parent 模型请求：若 parent 空闲，则在调用返回前追加消息；若 parent 正在准入或运行，则暂存报告，留到下一个安全日志位置。该模式不创建 inbox 条目实例，因此也不会产生虚构的继续执行管理器接受记录。
 
@@ -56,13 +56,13 @@ subagent seam 新增 `registerContinuableSetup(contribution): () => void`，由
 
 注册表负责注册、每个 child 的安装记录、设置回滚、child 作用域清理和立即撤销。应用一个批次会返回 Agent setup 提交对象，用于在所有 setup 的 await 均结算后、紧邻 Agent 发布前重新校验配置状态。因此，某项贡献抛出异常或被并发撤销时，会在 Agent 与会话发布前拒绝操作并回滚该批次。新注册项只会在驻留 child 的下一个 Activation 生效；移除注册项时，会先将它对新设置关闭，再立即撤销为正在配置或驻留的每个 child 安装的实例。注册 dispose（资源释放）与 child 上下文 dispose 都是幂等的，两者都会先尝试每项释放，再聚合失败。
 
-该 seam 使继续执行管理器无需知道工具名。report 包只安装 `report`；`@deepseek-ai/dsh-tool-subagent-control` 则独立安装 parent 侧的 `send_message` 和 `list_agents`。部署时可安装任一方向、同时安装两者或两者均不安装。提供方仍只负责数据，持久化描述符不会对 report 可用性或投递模式建立快照，冷恢复则使用部署当前的贡献与策略。
+该 seam 使继续执行管理器无需知道工具名。report 包只安装 `report` 及其 child 作用域指引 section；`@deepseek-ai/dsh-tool-subagent-control` 则独立安装 parent 侧的 `send_message` 和 `list_agents`。部署时可安装任一方向、同时安装两者或两者均不安装。提供方仍只负责数据，持久化描述符不会对 report 可用性或投递模式建立快照，冷恢复则使用部署当前的贡献与策略。
 
 ### 快照覆盖
 
-ACP（Agent Client Protocol）快照 harness 新增 `waitForSubagentTurnEnd`，按与 `session.N.jsonl` 相同的顺序选择第 N 个已收集 child。它会等待一个包含请求 header 的已闭合 child 轮次，以防可继续 child 早期播种描述符的轮次错误满足该边界。这样，整体组装的静默模式场景无需伪造 parent 可见信号，就能等待 child 侧报告。
+ACP（Agent Client Protocol）快照 harness 新增 `waitForSubagentTurnEnd`，按与 `session.N.jsonl` 相同的顺序选择第 N 个已收集 child。它会等待一个包含请求 header 的已闭合 child 轮次，以防可继续 child 早期播种描述符的轮次错误满足该边界。这样，整体组装的场景无需伪造 parent 可见信号，就能等待 child 侧报告。
 
-手写快照会启动一个可继续 child，执行真实的作用域局部 `report` 工具，确认空闲 parent 未被唤醒，然后提交一条后续 parent 提示词，使其消费封装后的报告。它声明 child schema pin `1`，因此本不属于全局的 `report` schema 会与 `tool-schemas.1.expected.json` 比对，root 则继续使用默认 schema pin。生成的工具目录会另外铸造一个 child 作用域，以收录同一个作用域局部 schema。
+手写快照会启动一个可继续 child，执行真实的作用域局部 `report` 工具，观察默认唤醒投递所产生的那一个普通 parent 轮次，然后提交一条后续 parent 提示词，使其消费封装后的报告。它声明 child pin `1`，因此本不属于全局的 `report` schema 与该 child 自身的提示词会分别与 `tool-schemas.1.expected.json` 和 `system-prompt.1.expected.md` 比对，root 则继续使用类别 pin。生成的工具目录会另外铸造一个 child 作用域，以收录同一个作用域局部 schema。
 
 ## 曾考虑的替代方案
 
@@ -72,7 +72,7 @@ ACP（Agent Client Protocol）快照 harness 新增 `waitForSubagentTurnEnd`，
 
 ### 始终唤醒 parent
 
-每次报告都唤醒 parent 会产生未经请求的轮次，还可能沿嵌套 subagent 级联扩散。静默投递更适合作为后台协调的默认值，而需要立即处理的部署可选择 wakeup。
+每次报告都唤醒 parent 会产生未经请求的轮次，还可能沿嵌套 subagent 级联扩散。当初选择静默投递作为默认值，前提是 parent 还有别的理由去读自己的上下文。[报告义务](2026-08-06-continuable-child-report-obligation.md)取代了该选择：已经停驻的后台协调者并没有这样的理由，因此唤醒成为默认值，而本段现在记录的是 `quiet` 为何仍然保留。
 
 ### 允许 child 选择投递模式
 
@@ -103,16 +103,16 @@ ACP（Agent Client Protocol）快照 harness 新增 `waitForSubagentTurnEnd`，
 - 只有安装 report 包贡献时，可继续进程内 child 才会恰好暴露一个作用域局部 `report` schema；无关 Agent 永远不会暴露该 schema。
 - 工具返回 parent 消息的稳定 `MessageId`。静默投递没有 `InboxItemId`；唤醒投递会产生一个单独的 inbox 条目实例。
 - 只有确切的驻留 child 才能报告，且只能报告给根据持久化谱系推导的确切在线直接 parent。服务不接受接收方参数，也不提供离线 fallback。
-- 静默投递是校验后的默认模式，绝不会启动 parent 请求。wakeup 会恰好创建一个后续 FIFO 轮次，绝不 steering 已开始的轮次。
+- 唤醒投递是校验后的默认模式：它会恰好创建一个后续 FIFO 轮次，绝不 steering 已开始的轮次。静默投递则绝不会启动 parent 请求。
 - parent 接受后取消或 dispose child 不会撤回报告。接受前，child dispose、drain、parent 丢失或调用方取消都会拒绝操作。
 - 新建和恢复的 Activation 都会在发布前组合当前设置贡献。新授权等待下一个 Activation 才生效，而已驻留 child 的授权撤销立即生效。
 - 单元覆盖固定可见性、allow-list 行为、两种投递模式、稳定的消息与发送方身份、嵌套路由、无效发送方、缺失的 parent、取消、drain、撤销竞争，以及不存在 Task 或隐式最终报告。
-- 无密钥整体组装快照证明真实 child 工具、静默且不唤醒的行为、持久化 parent 封装，以及 parent 后续消费。
+- 无密钥整体组装快照证明真实 child 工具、那一个被唤醒的 parent 轮次、持久化 parent 封装，以及 parent 后续消费。
 
 ### 已接受的风险
 
 该接受边界弱于持久化端到端投递。崩溃可能导致结果不明，重试则可能重复报告。
 
-wakeup 模式可能在嵌套 child 频繁报告时放大模型工作量。由部署所有者控制并默认静默，可以限制该风险，但无法完全消除。
+唤醒投递可能在嵌套 child 频繁报告时放大模型工作量。通过 `reportDelivery` 交由部署所有者控制，可以限制该风险，但无法完全消除。
 
 注册表中的存在性就是 parent 在线信号。宿主拥有的 parent 如果已开始 `AgentHandle.dispose()` 但尚未完成其作用域清理，仍可能接受并追加一条本进程不会再处理的报告。要弥合这个缺口，需要 Agent 层面的 dispose 开始信号，不能由 subagent 层推断。
diff --git a/.agents/notes/implemented/feature/2026-08-06-continuable-child-report-obligation.i18n.yaml b/.agents/notes/implemented/feature/2026-08-06-continuable-child-report-obligation.i18n.yaml
new file mode 100644
index 0000000000..652a4711f1
--- /dev/null
+++ b/.agents/notes/implemented/feature/2026-08-06-continuable-child-report-obligation.i18n.yaml
@@ -0,0 +1,6 @@
+# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
+# side as of the last confirmed-consistent state. Both languages carry equal authority;
+# after editing either side, bring the other along and re-record with:
+#   pnpm run verify-translation-pairing --write .agents/notes/implemented/feature/2026-08-06-continuable-child-report-obligation.md
+2026-08-06-continuable-child-report-obligation.md: f152ec1b8c353f094f2ba70785112eb1e165c510
+2026-08-06-continuable-child-report-obligation.zh.md: 4ec17e4642ffac385e6ce5464f41a3f4b3bebdbf
diff --git a/.agents/notes/implemented/feature/2026-08-06-continuable-child-report-obligation.md b/.agents/notes/implemented/feature/2026-08-06-continuable-child-report-obligation.md
new file mode 100644
index 0000000000..f152ec1b8c
--- /dev/null
+++ b/.agents/notes/implemented/feature/2026-08-06-continuable-child-report-obligation.md
@@ -0,0 +1,60 @@
+# Agent Note: The continuable child return channel is an obligation
+
+Status: implemented
+
+English | [中文](2026-08-06-continuable-child-report-obligation.zh.md)
+
+## Problem
+
+A continuable background child owns its own Session, so nothing it writes there reaches the agent that started it. [The report tool](2026-07-30-continuable-subagent-report-tool.md) gave that child a return channel and then presented it as one option among several: the schema said "call this zero or more times", nothing in the child's prompt asked it to call the tool at all, and the accepted default scheduling (`quiet`) added the report to a parked parent's next request without waking it.
+
+Each of those choices is defensible alone. Together they made the return channel unusable as a delegation contract. A child that finished its work, wrote its answer into its own transcript, and stopped left the parent with nothing; a child that did report reached a parent that had already parked and would not read the report until something unrelated woke it. External reports of parents busy-polling `list_agents`, re-sending messages to settled children, and abandoning `subagent` for `workflow` all reduce to the same missing guarantee.
+
+## Decision
+
+The return channel is an instruction the child receives, not a capability it may discover. The report package installs two scope-local registrations into every continuable in-process child, and one disposer revokes both:
+
+- the `report` tool, whose description now states that the child calls it once before finishing with a self-contained final result, and earlier for progress that changes what the parent should do next;
+- a `tool:report` system-prompt section at order 117 carrying the same obligation in the child's own voice, so a child that never reads tool descriptions closely still receives it.
+
+`reportDelivery` now defaults to `wakeup`. An accepted report creates exactly one ordinary later parent turn and wakes a parked parent driver; it still never steers an open turn. `quiet` remains available for deployments that prefer unread reports over turn amplification.
+
+### Why the section and the description both exist
+
+They address different failure modes. The tool description is read when the model is already considering `report`; the prompt section is read when it is deciding whether it is finished. The obligation belongs at both points because the failure this fixes — a child that simply stops — happens at the second one.
+
+The section is registered on the child's own scope, the same mechanism [child composition](../../../../packages/subagent/subagent/src/child-agent.ts) already uses for a shadowing persona, so the parent and every sibling see neither the tool nor the guidance. `installReportTool` rolls the section back if tool registration fails, and its returned disposer attempts both revocations before surfacing cleanup failures.
+
+### Instruction, not enforcement
+
+Nothing rejects a child that never reports. No runtime path inspects whether a report was sent, and `report` still accepts zero or many calls per turn. The change is model-facing wording plus a scheduling default; the service authority, acknowledgement, and recovery contracts are unchanged.
+
+That boundary is deliberate: prompt text can only reach a child that is still running its own loop. A child stopped by an error, a token ceiling, cancellation, or teardown never gets the chance to comply, which is why the runtime keeps its own account of settlement rather than trusting this instruction ([manager-owned settlement delivery](2026-08-06-manager-owned-subagent-settlement-delivery.md)).
+
+### Snapshot coverage
+
+The assembled ACP `subagent-report` scenario now exercises the shipped default: the child reports, the parked parent takes one ordinary turn on that report, and a later prompt still reads the report back out of the durable log. Because the child's scope now composes a prompt the class pin cannot describe, the snapshot harness gained `pinsChildSystemPrompts`, the exact counterpart of the existing `pinsChildToolSchemas`: it moves one child fixture's prompt into `system-prompt.<n>.expected.md`, leaves every other request-header field to the class pin, requires the sidecar exactly when declared, and rejects a sidecar identical to that class pin so a redundant copy cannot drift.
+
+## Alternatives considered
+
+**Keep `quiet` as the default and rely on the prompt alone.** This was the shipped position, and it supersedes nothing on its own: a report the parent never reads is indistinguishable from a report never sent. The [report-tool note's](2026-07-30-continuable-subagent-report-tool.md) rejection of always-waking assumed the parent had another reason to look at its context; a parked background coordinator does not. Turn amplification is the real cost, and it is now the reason `quiet` still exists rather than the reason it is the default.
+
+**Let the child choose the delivery mode per call.** Unchanged from the original rejection: the model would own scheduler pressure, and behavior would vary per call rather than per deployment.
+
+**Put the obligation only in the tool description.** A description is read while choosing among tools. The child this change targets is not choosing a tool; it believes it is done. Prompt guidance is the surface that reaches that decision.
+
+**Enforce the obligation at settlement by rejecting a silent child.** There is nothing to reject: by the time settlement is observable the child's loop is over, and failing its teardown would destroy work rather than deliver it. Delivering the terminal facts unconditionally from the runtime is the answer to that case, and it belongs to the continuation manager, not to this package.
+
+## Consequences
+
+- Every continuable in-process child with this package loaded carries one extra prompt section and a longer `report` description in every request; no other Agent's request changes.
+- The default deployment wakes the parent once per accepted report. A nested tree that reports frequently consumes extra parent turns; `quiet` is the documented escape.
+- `installReportTool` requires `ctx.systemPrompt` in the child scope, so the package declares `systemPrompt` in `inject` and fails at load rather than at the next child materialization.
+- Unit coverage pins the new default, two load-bearing instruction phrases, the section's child-only scope against both the parent and a sibling, and rollback or revocation of both registrations.
+- Three assembled ACP scenarios with continuable children pin the complete instruction text through the new sidecar; a future change to any child-scoped section fails those scenarios instead of passing silently.
+
+### Accepted risks
+
+Waking by default amplifies model work in deep trees. The deployment owns that through `reportDelivery`, and the amplification is bounded by one turn per accepted report.
+
+A child can still finish without reporting, and this change cannot detect it. Only the runtime's own [settlement account](2026-08-06-manager-owned-subagent-settlement-delivery.md) closes that case.
diff --git a/.agents/notes/implemented/feature/2026-08-06-continuable-child-report-obligation.zh.md b/.agents/notes/implemented/feature/2026-08-06-continuable-child-report-obligation.zh.md
new file mode 100644
index 0000000000..4ec17e4642
--- /dev/null
+++ b/.agents/notes/implemented/feature/2026-08-06-continuable-child-report-obligation.zh.md
@@ -0,0 +1,60 @@
+# Agent Note: 可继续 child 的返回通道是一项义务
+
+Status: implemented
+
+[English](2026-08-06-continuable-child-report-obligation.md) | 中文
+
+## 问题
+
+可继续后台 child 拥有自己的 Session，因此它写在那里的任何内容都不会到达启动它的 agent。[report 工具](2026-07-30-continuable-subagent-report-tool.md)为该 child 提供了一条返回通道，却把它呈现为若干选项之一：schema 里写着「可调用零次或多次」，child 的提示词中没有任何地方要求它调用该工具，而已采纳的默认调度（`quiet`）会把报告加入已停驻 parent 的下一次请求，却不唤醒它。
+
+这些选择单独看都站得住脚。合在一起，它们让这条返回通道无法作为委派契约使用。一个完成工作、把答案写进自己 transcript（文本记录）随后停止的 child，会让 parent 一无所获；而确实上报了的 child，面对的是一个已经停驻、要等到别的事件把它唤醒才会读到报告的 parent。外部反馈中的 parent 忙轮询 `list_agents`、反复向已结算 child 发送消息、以及放弃 `subagent` 改用 `workflow`，都可归结为同一处缺失的保证。
+
+## 决策
+
+返回通道是 child 收到的一条指令，而不是它需要自行发现的能力。report 包会向每个可继续进程内 child 安装两项作用域局部注册，并由同一个 disposer 撤销两者：
+
+- `report` 工具，其描述现在说明 child 要在结束前调用一次并给出自足的最终结果，并在部分进展会改变 parent 下一步动作时提前调用；
+- 一个 order 为 117 的 `tool:report` 系统提示词 section，用 child 自己的语气承载同一条义务，使从不细读工具描述的 child 仍能收到它。
+
+`reportDelivery` 的默认值现在是 `wakeup`。一条被接受的报告恰好创建一个普通的后续 parent 轮次并唤醒停驻的 parent 驱动；它仍然绝不 steering（中途引导）已开始的轮次。对于宁可让报告无人阅读也要避免轮次放大的部署，`quiet` 依旧可用。
+
+### 为什么 section 与描述同时存在
+
+两者针对不同的失效模式。工具描述是在模型已经在考虑 `report` 时被读到的；提示词 section 是在它判断自己是否已经完成时被读到的。这条义务必须同时出现在两处，因为本次修复的失效——child 直接停下——发生在第二处。
+
+该 section 注册在 child 自己的作用域上，与[child 组合](../../../../packages/subagent/subagent/src/child-agent.ts)为遮蔽式 persona 已经使用的机制相同，因此 parent 与所有同级都看不到该工具与该指引。工具注册失败时，`installReportTool` 会回滚该 section；它返回的 disposer 会先尝试撤销两项注册，再抛出清理失败。
+
+### 是指令，不是强制
+
+没有任何东西会拒绝一个从不上报的 child。没有任何运行时路径会检查是否发送过报告，`report` 仍接受一个轮次中调用零次或多次。本次改动是面向模型的措辞加上一个调度默认值；服务权限、确认与恢复契约都保持不变。
+
+这条边界是刻意划定的：提示词文本只能到达仍在运行自身循环的 child。被错误、token 上限、取消或拆卸终止的 child 根本没有机会遵守，因此运行时会自己记录结算这件事，而不是信任这条指令（见[由管理器负责的结算投递](2026-08-06-manager-owned-subagent-settlement-delivery.md)）。
+
+### 快照覆盖
+
+整体组装的 ACP `subagent-report` 场景现在演练随附的默认行为：child 上报，停驻的 parent 就该报告执行一个普通轮次，随后的提示词仍能从持久化日志中把报告读回来。由于该 child 的作用域现在组合出类别 pin 无法描述的提示词，快照 harness 新增了 `pinsChildSystemPrompts`，它与既有 `pinsChildToolSchemas` 完全对称：把一个 child fixture 的提示词移入 `system-prompt.<n>.expected.md`，其余请求 header 字段仍归类别 pin 所有，要求 sidecar 恰好在声明时存在，并拒绝与该类别 pin 完全相同的 sidecar，使冗余副本无法悄悄漂移。
+
+## 备选方案
+
+**保留 `quiet` 作为默认值，只依赖提示词。** 这曾是随附的立场，而它本身什么也没有解决：一条 parent 从不阅读的报告，与一条从未发送的报告无法区分。[report 工具 Agent Note](2026-07-30-continuable-subagent-report-tool.md)对「始终唤醒」的否决，前提是 parent 还有别的理由去查看自己的上下文；已停驻的后台协调者并没有。轮次放大才是真正的代价，而它现在是 `quiet` 仍然保留的理由，而不是它作为默认值的理由。
+
+**让 child 按调用选择投递模式。** 与最初的否决相同：模型将掌握调度压力，行为也会随调用而非随部署变化。
+
+**只把义务写在工具描述里。** 描述是在从多个工具中选择时被读到的。本次改动针对的 child 并不在选择工具，它认为自己已经做完了。提示词指引才是能触及该判断的界面。
+
+**在结算时拒绝沉默的 child，以此强制该义务。** 没有什么可以拒绝：当结算可被观察时 child 的循环已经结束，让它的拆卸失败只会毁掉工作而不会送达结果。由运行时无条件投递终止事实才是这一情形的答案，而它属于继续执行管理器，不属于本包。
+
+## 后果
+
+- 加载本包后，每个可继续进程内 child 的每次请求都会多出一个提示词 section 和一段更长的 `report` 描述；其他任何 Agent 的请求都不变。
+- 默认部署会为每条被接受的报告唤醒 parent 一次。频繁上报的嵌套树会消耗额外的 parent 轮次；`quiet` 是有文档记载的退路。
+- `installReportTool` 需要 child 作用域中的 `ctx.systemPrompt`，因此本包在 `inject` 中声明 `systemPrompt`，从而在加载时失败，而不是等到下一次 child 物化时。
+- 单元覆盖固定了新默认值、两处关键指令措辞、该 section 相对 parent 与同级均仅限 child 的作用域，以及两项注册在安装回滚或撤销时的清理。
+- 三个带可继续 child 的整体组装 ACP 场景通过新的 sidecar 逐字固定完整的 child 提示词；今后任何对 child 作用域 section 的改动都会让这些场景失败，而不是悄悄通过。
+
+### 已接受的风险
+
+默认唤醒会在深层树中放大模型工作量。部署通过 `reportDelivery` 掌握该取舍，且放大幅度以每条被接受报告一个轮次为界。
+
+child 仍可能不上报就结束，本次改动无法检测这一点。只有运行时自己的[结算记账](2026-08-06-manager-owned-subagent-settlement-delivery.md)才能补上这一情形。
diff --git a/.agents/notes/implemented/feature/2026-08-06-manager-owned-subagent-settlement-delivery.i18n.yaml b/.agents/notes/implemented/feature/2026-08-06-manager-owned-subagent-settlement-delivery.i18n.yaml
new file mode 100644
index 0000000000..c709af4080
--- /dev/null
+++ b/.agents/notes/implemented/feature/2026-08-06-manager-owned-subagent-settlement-delivery.i18n.yaml
@@ -0,0 +1,6 @@
+# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
+# side as of the last confirmed-consistent state. Both languages carry equal authority;
+# after editing either side, bring the other along and re-record with:
+#   pnpm run verify-translation-pairing --write .agents/notes/implemented/feature/2026-08-06-manager-owned-subagent-settlement-delivery.md
+2026-08-06-manager-owned-subagent-settlement-delivery.md: e06fe2b4ca7dd9a975f70524979de09d13cdcf79
+2026-08-06-manager-owned-subagent-settlement-delivery.zh.md: 77e8b5e4bae9b05e08bf2c4a4bf997b688cd1157
diff --git a/.agents/notes/implemented/feature/2026-08-06-manager-owned-subagent-settlement-delivery.md b/.agents/notes/implemented/feature/2026-08-06-manager-owned-subagent-settlement-delivery.md
new file mode 100644
index 0000000000..e06fe2b4ca
--- /dev/null
+++ b/.agents/notes/implemented/feature/2026-08-06-manager-owned-subagent-settlement-delivery.md
@@ -0,0 +1,103 @@
+# Agent Note: Settlement delivery belongs to the continuation manager
+
+Status: implemented
+
+English | [中文](2026-08-06-manager-owned-subagent-settlement-delivery.zh.md)
+
+## Problem
+
+Continuable background delegation was the one asynchronous operation a model could start but could not reach the end of. Every other shape has a retrieval primitive or a return value: a background bash command and a one-shot background subagent both settle through a Task that `task_output(wait: true)` can block on, a workflow and a foreground subagent return their result to the caller. A continuable background child returned only its durable id, and nothing existed that a parent could wait on or would be handed.
+
+[The report obligation](2026-08-06-continuable-child-report-obligation.md) closed the cooperative half of that gap by instructing the child to report before it finishes. Instruction cannot close the rest. A child stopped by a token ceiling, a model failure, cancellation, or teardown never reaches the point where it could comply — not rarely, but never — and those are precisely the endings a waiting parent most needs to hear about. The observable downstream symptoms were parents busy-polling `list_agents`, re-sending messages to children that had already settled, and deployments abandoning `subagent` for `workflow` because a workflow at least returns something.
+
+The signal already existed. `subagent/end` has carried `stopReason` and `lastAssistantMessage` since continuable Activations shipped. What was missing was any consumer that turned it into context the parent's model could see.
+
+## Decision
+
+The continuation manager delivers the account itself, from inside the disposal transaction that ends the Activation.
+
+When a resident Activation settles, `notifySettlement()` resolves the child's durable direct parent and sends it one user-role message: the epoch's outcome as a sentence the parent can act on, then the child's final assistant content, or a statement that it produced none. Delivery is unconditional for every child whose id a caller actually received. It does not consult whether the child reported, and it keeps no bookkeeping that could make the promise conditional — that unconditionality is what lets `tool-subagent` tell the model "you are told when it finishes, so never poll or wait on it" and have that be true. A materialization rolled back before its first accepted message stays silent, because the caller was told that child was not established.
+
+### Provenance
+
+The notice carries `{ kind: 'subagent-settled', form: 'notice', summary, senderSessionId }`. It is deliberately not the existing `subagent-report` kind. A report is content the child chose; this is the runtime stating what became of the child. Merging them would credit the child with words it never wrote, and would make a durable log unable to distinguish "the child said it was done" from "the harness observed that it stopped". The `notice` form also gives a UI the collapsed one-line presentation this message wants, where `relay` would present it as correspondence.
+
+### Two ordering rules, and why the manager owns them
+
+An external `ctx.on('subagent/end')` listener looks more decoupled and is wrong. `SubagentRunEndInfo` names no parent, the child handle is already disposed when the edge fires so the parent cannot be recovered from it, and the ownership release that wakes the parent's own settlement watcher has already run. The manager holds the parent reference throughout disposal, so none of those obstacles exist for it.
+
+**The send happens before `releaseOwnership`.** At that point the parent still counts this child, so `stateOf(parent)` is `waiting` and the parent is structurally unable to be judged settled. Delivering after the release instead races a watcher that resumes one microtask later, finds itself childless and quiet, and disposes an Agent whose `cancel()` clears the very inbox the notice is sitting in. The failure mode is a silently missing message with no error anywhere.
+
+**A resident parent receives it through `admitWaking`.** Registering the message id before the synchronous send is what keeps the window between `followup()` and the microtask that admits it from being read as quiescence. This is not belt-and-braces over the first rule: `Agent.status` folds context maintenance into `idle`, and a waking send behind maintenance only arms a deferred wake, so a parent compacting its context is judged quiet by both `status` and the owned-child set the moment the release lands.
+
+Both rules are pinned by tests that fail when the ordering is reversed or the accounting removed.
+
+### Scheduling
+
+An idle parent gets one ordinary later turn. A busy parent is steered into its nearest step boundary, because `Inbox.claim()` takes the whole next-step batch at one boundary: four children settling together then cost one step rather than four turns. Steering rather than injecting is deliberate — the wake is a no-op while the driver is running, and it closes the window where a driver retires between the status read and the send, which would strand the notice unclaimed until something unrelated woke the parent. This is a correctness rule, not a deployment preference, so it is not a `Config` field.
+
+One `running` parent is not steerable: one whose turn is already cancelled but has not yet exited. `Agent.send()` redirects waking input submitted after cancellation to the next turn, latches the wake, and replays it once the cancelled driver converges — except for a disposal cancellation, which never latches and belongs to the teardown rule below. The notice therefore still opens its own turn without waiting for unrelated input; the cost is a redirected turn boundary, not the message.
+
+**A parent whose own teardown began gets no wake.** Waking is not a queue operation: `Agent.followup()` on a quiescent Agent starts a turn, and `cancel()` on an idle Agent is a documented no-op that does not arm against a later one. Every teardown path therefore ends with a live, cancelled, still-registered parent — `drainContinuableDescendants()` is called by the ACP bridge between cancelling its session agents and disposing them — so an unguarded notice starts a real model request on an Agent about to be destroyed, once per tree layer, because each layer's own notice then wakes the layer above it. `notifySettlement()` asks the same question `assertAdmitting()` asks (is this lineage's continuable admission closed?) and injects instead. Injection is not a durable mailbox — Accepted risks records what the parent's own disposal then does to it — but it is the only send that reaches a parent still reading its inbox without arming a turn on one that is not, and nothing is lost that the wake would have delivered: the turn a wake started was itself disposed mid-flight.
+
+Delivery never blocks or fails teardown. A rejected send is logged and dropped, because retaining a child to retry a notice would pin its whole ancestry in `waiting` forever, and a parent that has left the registry is an ordinary outcome rather than an error.
+
+### The epoch's own log is the whole account
+
+`epochStopReason()` reads the epoch's outcome from its own log, because teardown succeeding says nothing about whether the model errored, hit its ceiling, or was stopped. Reading turns alone got that wrong twice, in the same shape both times: a turn stopped before its first step leaves a `turn/end` indistinguishable from the balanced no-op turns a rejection or an emptied claim produces, so the filter that skipped those also skipped real endings and answered with the previous turn's clean completion. The durability checkpoint (`dsh-session-checkpoint-policy`, in every shipped profile) and prompt assembly both run at that boundary and both propagate, and `Inbox.claim()` has already taken the messages by then — so the parent was told a child finished while the delivery it was waiting on had been swallowed. Under a promise that says "you are told when it finishes, so never poll", that is the one failure a parent cannot detect and will not retry.
+
+The missing fact was never the turn's; it was the inbox's. `Inbox` logs every mutation with `removedCount` and marks a cancellation `outcome: 'canceled'`, which separates a turn claiming its input from work being dropped unrun. `foldConsumedWork()` in `dsh-agent` folds both vocabularies into one answer: the latest turn that accounts for consumed work — stepped, or claimed-then-failed, stopped, or rejected — and whether accepted work was cancelled after it with no turn opening over it. A `blocked` end over claimed input is an account too: the pre-step rejection that produced it — a hook deny, a policy plugin — discarded the messages the turn claimed, so the notice says the child declined rather than finished. Only a `blocked` turn that claimed nothing stays invisible.
+
+Deriving it from the log rather than from live state is what makes it whole. An earlier version sampled the manager's own Activation immediately before cancelling, which could only ever see cancellations this manager was about to perform: an ancestor's `interrupt()`, or an unloading plugin cancelling an agent it tracks, left the sample false and the notice still saying `finished`. It also left the accepted-but-never-claimed case pinned to nothing a test could distinguish from its absence. One fold over the log covers every issuer, and both halves fail their own tests when removed.
+
+Precedence is the consumer's: a recorded failure or ceiling wins over a cancellation, because stopping a child that had already failed does not turn its failure into a cancellation. `dsh-agent` owns the fold because it owns the inbox marker the answer depends on, and both consumers already depend on it — the continuable epoch here, and the one-shot `readResult()`, which had the same hole.
+
+Both matter past the notice: `subagent/end` carries `stopReason` to the jsonrpc UI and the Claude hook bridge, which reported a torn-down mid-turn child as `completed`.
+
+### Snapshot coverage
+
+Three assembled ACP scenarios cover the notice: a child that never reports, a child that reports first, and a child driven through several follow-up turns. All three needed an explicit fence. The notice arrives once the child's teardown finishes, which races whatever the parent is already doing, so each scenario holds the child behind the parent's spawn turn and then waits for the parent turn the notice opens (`waitForTurnStart` at that turn, then `waitForTurnEnd`) before the script continues. Waiting for a turn the run is not fenced to produce is not coverage: it is a timeout when the notice lands in the turn already running instead.
+
+`subagent-continuable` is the one that pins a failure. Its child's last turn dies on the forced durability checkpoint without entering a step, so that transcript is where the stop-reason rule above is visible end to end: the notice says the child *failed*, carries the earlier `SECOND_OK` as its last content rather than as a result, and the parent's own acknowledgement turn reaches the ACP client.
+
+A keyless headless Loader snapshot covers the user-visible path end to end. Its replay parent starts one continuable child with `run_in_background: true`, never calls `list_agents`, `send_message`, or Task tools, consumes the manager-authored `subagent-settled` notice, and produces its final answer. The child never calls `report`, so the transcript cannot pass through the cooperative report path. A test-only Loader fence holds the parent's post-spawn request until the real manager notice enters its inbox, removing platform scheduling from the transcript without synthesizing the notice.
+
+`subagent-report` needed one more concession. With the shipped waking report default, that scenario has two independent parent wakes — the report and the settlement — and whether the second extends the first's turn or opens its own is a genuine coin flip that measured 50/50 across runs. No authored transcript can hold both orders. Its overlay therefore pins `reportDelivery: quiet`, leaving settlement as the only wake, and a snapshot-only pre-step fence holds the child until the parent's spawn turn ends so that wake opens one deterministic turn claiming both messages. The waking report default keeps its coverage in the report package's own tests.
+
+The refusal and interruption wordings are pinned verbatim in unit tests rather than in a replayed transcript: producing them needs a rejecting policy plugin or a cancellation fenced at a step boundary, which the keyless assemblies do not otherwise carry, and the assembled scenarios already pin the notice pathway itself end to end.
+
+## Alternatives considered
+
+**Give continuable children a Task.** A Task is a one-shot contract: one producer, one settlement, one result. An Activation runs many turns, outlives any single one, and can be resumed after it ends. Wrapping it in a Task recreates exactly the lifetime mismatch continuable children were introduced to remove, and would make one turn look terminal.
+
+**Attach an external `subagent/end` listener.** Rejected on three counts above — no parent in the payload, a disposed child handle, and an ordering the listener cannot influence. A listener would also have to be strictly synchronous to beat the release, and nothing at that seam enforces it, so the correct version would be correct only by accident.
+
+**Deliver only when the child did not report.** This was the first design. It needs per-Activation bookkeeping, still misses the child that reported progress and then died before its result, and — decisively — makes the parent-facing promise conditional. "Usually you are told" is not a contract a tool description can state, and a model that cannot rely on the notice will poll anyway.
+
+**Make delivery configurable.** A deployment switch would return the model-facing text to "usually", which is the failure this change exists to remove. Protocol constants and safety invariants stay fixed; this is one of them.
+
+**Change `subagent/end` to carry the parent, and let a plugin deliver.** That widens a published payload for one in-package consumer, keeps every ordering hazard, and makes the return channel an optional plugin again. Extending the package-private `ActivationObserver` with `terminal(failure)` keeps one computation of the terminal facts and no public surface change.
+
+**Always use `followup`.** Simpler and uniform, but a fan-out of children settling together would cost one parent turn each. The step-boundary batch already exists; using it is free.
+
+## Consequences
+
+- A continuable child's parent receives one message per settled Activation. Fan-out deployments therefore add parent turns; steering keeps a simultaneous batch to one step.
+- `tool-subagent` promises the notice in its schema because the return channel is service behavior, not an optional plugin.
+- `Activation` carries `parentSession` and `announced`. The first exists because the child handle is disposed before delivery; the second is what keeps a rolled-back materialization silent.
+- `foldConsumedWork()` replaces `dsh-session`'s `findLastMessageTurnEnd()` and moves to `dsh-agent`, which owns the inbox marker it reads; the one-shot in-process path folds the same answer and does not classify a cut-short one-shot child as `completed`.
+- Unit coverage pins the unconditional contract, each terminal reason, idle and busy scheduling, the batch, the maintenance regression, the pre-release ordering, a parent that is gone, and a rejected send that must not fail teardown.
+- Three ACP scenarios use an explicit settlement fence, and `subagent-report` has a config overlay that pins quiet report delivery.
+- A keyless headless Loader snapshot pins background start → manager-authored settlement notice → final parent answer with no polling or child `report` call.
+
+### Accepted risks
+
+The notice is delivered, not confirmed. There is no durable mailbox, receipt, or retry: a parent that is not live loses it, and the child's Session remains the only durable record. Closing that needs an offline mailbox protocol with its own addressing, authorization, and replay rules.
+
+A notice injected during teardown is not read by a model when that parent is disposed next, which every teardown caller does: the disposal cancel clears the unclaimed message and the log keeps the insert/cancel pair as the record. Making teardown delivery readable after resume requires either the offline mailbox above or a change to disposal of durable pending work. Disposal discards every unclaimed inbox item, including user input, so changing that behavior is a core-agent decision rather than a settlement-delivery detail. After resume, the parent can discover the child but does not receive the outcome: `list_agents` reports existence and live-or-stored status only — `SubagentListEntry.activity` says so — and recovering the ending requires asking the child through `send_message`.
+
+Stop-reason attribution is a best effort over the log's existing splice vocabulary, biased against overstating success. `Inbox.remove()` and teardown's `clear()` write identical cancellation splices, so removing a message whose content survives elsewhere — `workspace-context` vacuuming a pending instruction refresh, or settlement's own cancel clearing one left pending — can read as work dropped unrun and report a finished child as stopped. Separating them requires a richer removal vocabulary in `dsh-agent`; without it, the misread is narrow and errs toward the parent double-checking a finished child, never toward trusting an unfinished one.
+
+Turn amplification is real for deep or wide trees, and it is not configurable by design. The step-boundary batch bounds it for simultaneous settlement but not for children that settle apart.
+
+Two independent waking sources cannot be ordered in an authored transcript. The assembled coverage pins each separately rather than their interleaving.
diff --git a/.agents/notes/implemented/feature/2026-08-06-manager-owned-subagent-settlement-delivery.zh.md b/.agents/notes/implemented/feature/2026-08-06-manager-owned-subagent-settlement-delivery.zh.md
new file mode 100644
index 0000000000..77e8b5e4ba
--- /dev/null
+++ b/.agents/notes/implemented/feature/2026-08-06-manager-owned-subagent-settlement-delivery.zh.md
@@ -0,0 +1,103 @@
+# Agent Note: Settlement delivery belongs to the continuation manager
+
+Status: implemented
+
+[English](2026-08-06-manager-owned-subagent-settlement-delivery.md) | 中文
+
+## Problem
+
+可继续后台委派是模型唯一一种能够发起、却无法抵达终点的异步操作。其他每一种形态都有取回原语或返回值：后台 bash 命令与一次性后台 subagent 都通过 Task 结算，`task_output(wait: true)` 可以阻塞等待；workflow 与前台 subagent 会把结果返回给调用方。可继续后台 child 只返回它持久化的 id，而父级既没有可等待的对象，也不会被交付任何东西。
+
+[报告义务](2026-08-06-continuable-child-report-obligation.md)通过要求 child 在结束前上报，补上了这一缺口中协作的那一半。指令无法补上其余部分。被 token 上限、模型失败、取消或拆卸终止的 child 永远走不到能够遵守的那一步——不是很少，而是从不——而这些恰恰是等待中的父级最需要被告知的结束方式。可观察到的下游症状包括：父级忙轮询 `list_agents`、向已经结算的 child 反复发送消息，以及部署放弃 `subagent` 转用 `workflow`，因为 workflow 至少会返回点什么。
+
+信号本身早就存在。自可继续 Activation 发布以来，`subagent/end` 就一直携带 `stopReason` 与 `lastAssistantMessage`。缺的是把它变成父级模型能看到的上下文的那个消费者。
+
+## Decision
+
+继续执行管理器自己投递这份记账，就在结束 Activation 的那笔 dispose 事务内部完成。
+
+当驻留 Activation 结算时，`notifySettlement()` 解析该 child 持久化的直接父级，并向它发送一条用户角色消息：先是父级可据以行动的一句结果说明，然后是 child 的最终 assistant 内容，或一句说明它没有产出内容。对每个调用方真正拿到过 id 的 child，投递都是无条件的。它不查询 child 是否上报过，也不保留任何可能让这项承诺变成有条件的记账——正是这种无条件性，才让 `tool-subagent` 能够告诉模型「它结束时你会被告知，因此绝不要轮询或等待它」并且这句话为真。在第一条消息被接受之前就回滚的物化保持静默，因为调用方已被告知该 child 未建立。
+
+### 来源信息
+
+该通知携带 `{ kind: 'subagent-settled', form: 'notice', summary, senderSessionId }`，刻意不复用既有的 `subagent-report` kind。上报是 child 选择的内容；这条消息则是运行时在陈述这个 child 后来怎样了。把两者合并会把 child 从未写过的话算到它头上，也会让持久化日志无法区分「child 说它做完了」和「harness 观察到它停下了」。`notice` 形态还为 UI 提供了这条消息想要的折叠单行呈现，而 `relay` 会把它呈现为往来信件。
+
+### 两条顺序规则，以及为什么归管理器所有
+
+外部 `ctx.on('subagent/end')` listener 看起来更解耦，但它是错的。`SubagentRunEndInfo` 不指名父级；该边触发时 child handle 已被 dispose，因此无法从中恢复父级；而唤醒父级自身结算 watcher 的所有权释放也已经执行过了。管理器在整个 dispose 过程中都持有父级引用，因此这些障碍对它都不存在。
+
+**发送发生在 `releaseOwnership` 之前。** 此刻父级仍然计入这个 child，因此 `stateOf(parent)` 为 `waiting`，父级在结构上不可能被判定为已结算。改在释放之后投递，则会与一个在下一个 microtask 恢复的 watcher 竞争：它会发现自己没有 child 且处于静止，于是 dispose 一个 Agent，而该 Agent 的 `cancel()` 会清空正装着这条通知的那个 inbox。失效表现是一条静默丢失的消息，任何地方都不会报错。
+
+**驻留父级通过 `admitWaking` 接收它。** 在同步发送之前登记消息 id，正是让 `followup()` 与承认它的那个 microtask 之间的窗口不被读作静止的原因。这不是对第一条规则的多余保险：`Agent.status` 会把上下文维护折叠成 `idle`，而维护期间的唤醒发送只会预置一次延后唤醒，因此正在压缩上下文的父级，在所有权释放落地的那一刻会同时被 `status` 与已拥有 child 集合判定为静止。
+
+两条规则都有测试固定：把顺序反转或去掉记账，测试就会失败。
+
+### 调度
+
+空闲父级得到一个普通的后续轮次。繁忙父级则被 steer 到其最近的 step 边界，因为 `Inbox.claim()` 会在一个边界上整批取走 next-step：四个 child 同时结算时因此只消耗一个 step，而不是四个轮次。采用 steer 而非 inject 是刻意的——驱动运行期间该唤醒是空操作，同时它关闭了「驱动在状态读取与发送之间退出」的那个窗口；否则通知会滞留无人认领，直到别的事件唤醒父级。这是正确性规则而非部署偏好，因此不做成 `Config` 字段。
+
+有一种 `running` 父级是无法 steer 的：轮次已被 cancel 但尚未退出的那种。`Agent.send()` 会把取消之后提交的唤醒输入改投到下一个轮次、闩存这次唤醒，并在被取消的驱动收敛后重放它——只有 disposal 取消从不闩存，那属于下面的拆卸规则。因此通知仍会开启自己的轮次，无需等待无关输入；代价是一次被改投的轮次边界，而不是消息本身。
+
+**自身已开始拆卸的父级不会被唤醒。** 唤醒不是入队操作：对静息 Agent 调用 `Agent.followup()` 会开启一个轮次，而对空闲 Agent 调用 `cancel()` 是文档明确的空操作，不会对之后的轮次设防。因此每条拆卸路径最终都面对一个在线、已取消、仍在注册表中的父级——ACP 桥接层正是在取消其 session agent 与 dispose 它们之间调用 `drainContinuableDescendants()`——于是一条无防护的通知会在一个即将被销毁的 Agent 上发起真实模型请求，而且每层树各一次，因为每层自己的通知又会唤醒它上面那层。`notifySettlement()` 会问 `assertAdmitting()` 问的同一个问题（这条谱系的可继续准入是否已关闭？），并改为 inject。inject 不是持久 mailbox——父级自身的 dispose 会对它做什么，记在「已接受的风险」里——但它是唯一能送达仍在读取自身 inbox 的父级、又不会在不该被唤醒的父级上预置一个轮次的发送方式；而唤醒本可送达的东西一样没有丢失：唤醒开启的那个轮次本身就会在半途被 dispose。
+
+投递绝不会阻塞或使拆卸失败。发送被拒会被记录并丢弃，因为为重试一条通知而保留 child，会把它的整条祖先链永久钉在 `waiting` 上；而父级已离开注册表属于普通结果，不是错误。
+
+### epoch 自己的日志就是全部交代
+
+`epochStopReason()` 从 epoch 自己的日志读取结局，因为拆卸成功与否，对「模型是否报错、是否撞到上限、是否被停下」什么也没说明。只读轮次这件事已经错了两次，而两次的形状相同：在第一个 step 之前被停下的轮次，其 `turn/end` 与「拒绝」或「被清空的认领」产生的平衡空转轮次长得一模一样，于是那道用来跳过后者的过滤，也把真实的结局一起跳过了，转而用上一个轮次的干净收尾作答。持久化检查点（`dsh-session-checkpoint-policy`，存在于每个随附 profile 中）与提示词组装都运行在这个边界上、且都会向外传播，而此时 `Inbox.claim()` 已经把消息取走了——于是父级被告知 child 已完成，而它正在等待的那条投递已被吞掉。在「你会在它完成时被告知，所以永远不要轮询」这一承诺之下，这恰恰是父级无法察觉、也不会重试的那一种失败。
+
+缺失的事实从来不属于轮次，而属于 inbox。`Inbox` 会把每次改动连同 `removedCount` 一起记入日志，并给取消标记 `outcome: 'canceled'`，这就把「某个轮次认领了它的输入」与「工作被丢弃且从未运行」区分开来。`dsh-agent` 中的 `foldConsumedWork()` 把两套词汇折叠成一个答案：能为已消费工作作出交代的最新轮次——进入过 step 的，或认领后失败、被停下或被拒绝的——以及此后是否有已接受的工作被取消、而没有任何轮次为它开启过。认领过输入、以 `blocked` 结束的轮次同样是一份交代：产生它的 pre-step 拒绝——hook deny、策略插件——把该轮次认领的消息一并丢弃了，因此通知会说 child 拒绝了任务，而不是完成了任务。只有没认领任何输入的 `blocked` 轮次保持不可见。
+
+从日志而不是从活动状态推导，才让它完整。早先的版本会在 cancel 之前立刻采样管理器自己的 Activation，而那样只能看到本管理器即将执行的取消：来自祖先的 `interrupt()`，或某个正在卸载的插件取消它所跟踪的 Agent，都会让该采样为假，通知照旧说 `finished`。它也让「已接受但从未被认领」这一情形没有任何测试能把它与「该判据不存在」区分开。一次对日志的折叠覆盖了所有发起方，而两个半边在被移除时都会让各自的测试失败。
+
+优先级归消费方：已记录的失败或上限优先于取消，因为停下一个已经失败的 child，不会把它的失败变成一次取消。`dsh-agent` 拥有这个 fold，是因为答案所依赖的那个 inbox 标记归它所有，而两个消费方本来就依赖它——这里的可继续 epoch，以及一次性的 `readResult()`（它有同一个漏洞）。
+
+两者的影响都超出通知本身：`subagent/end` 会把 `stopReason` 送到 jsonrpc UI 与 Claude hook 桥接层，而它们此前把被拆卸的、正在跑轮次的 child 报成 `completed`。
+
+### 快照覆盖
+
+三个整体组装的 ACP 场景覆盖该通知：一个从不上报的 child、一个先上报的 child，以及一个被多轮 follow-up 驱动的 child。三者都需要显式栅栏。通知在 child 拆卸完成后才到达，会与父级当时正在做的事竞争，因此每个场景都会把 child 保持到父级启动轮次结束，再等待该通知开启的那个父级轮次（先 `waitForTurnStart` 到该轮次，再 `waitForTurnEnd`），然后脚本才继续。等待一个运行并未被栅栏保证会产生的轮次不算覆盖：一旦通知落进已经在跑的那个轮次，它就是一次超时。
+
+`subagent-continuable` 是其中固定失败结局的那个。它的 child 最后一个轮次在被强制的持久化检查点上死亡，且未进入任何 step，因此该 transcript 正是上面那条终止原因规则的端到端可见之处：通知说该 child **失败**，把此前的 `SECOND_OK` 作为它最后产出的内容而非结果携带，而父级自己的确认轮次会到达 ACP 客户端。
+
+另有一个无密钥的 headless Loader 快照端到端覆盖用户可见路径。其重放父级通过 `run_in_background: true` 启动一个可继续 child，从不调用 `list_agents`、`send_message` 或 Task 工具，消费管理器写入的 `subagent-settled` 通知，并给出最终答案。child 从不调用 `report`，因此该 transcript 不可能经由协作式上报路径通过。一个仅用于测试的 Loader 栅栏会把父级启动后的请求保持到真实管理器通知进入其 inbox 为止，从 transcript 中排除平台调度差异，但不会伪造该通知。
+
+`subagent-report` 还需要多做一步让步。在随附的唤醒上报默认值下，该场景有两个互相独立的父级唤醒——上报与结算——而第二个究竟是延长第一个的轮次还是另开一个轮次，是一枚真正的硬币，多次运行实测约为五五开。任何手写 transcript 都无法同时容纳两种顺序。因此它的 overlay 固定 `reportDelivery: quiet`，使结算成为唯一唤醒；另一个仅用于快照的 pre-step 栅栏会把 child 保持到父级启动轮次结束，使这次唤醒开启一个确定轮次并同时认领两条消息。唤醒上报默认值的覆盖则保留在 report 包自身的测试中。
+
+拒绝与中断两种措辞在单元测试中逐字钉死，而不进入重放 transcript：触发它们需要一个会拒绝的策略插件、或一次在 step 边界被栅栏卡住的取消，而无密钥组装本身并不携带这些；通知通路本身已由整体组装场景端到端钉住。
+
+## Alternatives considered
+
+**给可继续 child 引入 Task。** Task 是一次性契约：一个生产者、一次结算、一个结果。Activation 会执行许多轮次、比其中任何一轮活得更久，并且可以在结束后被恢复。用 Task 包装它，恰好重建了可继续 child 当初为消除而引入的生命周期错配，还会让某一个轮次看起来是终局。
+
+**挂一个外部 `subagent/end` listener。** 因上文三点被否决——payload 里没有父级、child handle 已被 dispose，以及 listener 无法影响的顺序。listener 还必须严格同步才能抢在释放之前，而该 seam 上没有任何东西强制这一点，因此正确的版本只能靠碰巧正确。
+
+**仅在 child 没有上报时投递。** 这是最初的设计。它需要按 Activation 记账，仍会漏掉「报了进度、随后在给出结果前死掉」的 child，而且最关键的是：它让面向父级的承诺变成有条件的。「通常你会被告知」不是工具描述能陈述的契约，而无法依赖该通知的模型无论如何都会去轮询。
+
+**把投递做成可配置。** 部署开关会把面向模型的文本重新变回「通常」，而这正是本次改动要消除的失效。协议常量与安全不变量保持固定；这就是其中之一。
+
+**修改 `subagent/end` 让它携带父级，由插件负责投递。** 那会为一个包内消费者拓宽已发布的 payload，保留全部顺序风险，并让返回通道重新变成可选插件。以 `terminal(failure)` 扩展包私有的 `ActivationObserver`，则只保留一处终止事实的计算，且不改动任何公开面。
+
+**始终使用 `followup`。** 更简单也更统一，但一批同时结算的 child 会各自消耗一个父级轮次。step 边界的批量语义本来就存在，用它是免费的。
+
+## Consequences
+
+- 可继续 child 的父级会为每个已结算 Activation 收到一条消息。因此，做扇出的部署会增加父级轮次；steer 会把同时结算的一批压缩到一个 step。
+- `tool-subagent` 在其 schema 中承诺该通知，因为返回通道是服务行为，不是可选插件。
+- `Activation` 携带 `parentSession` 与 `announced`。前者存在是因为 child handle 在投递前已被 dispose；后者让被回滚的物化保持静默。
+- `foldConsumedWork()` 取代 `dsh-session` 的 `findLastMessageTurnEnd()`，并迁移到 `dsh-agent`——它拥有该 fold 所读取的 inbox 标记；一次性 in-process 路径折叠同一个答案，不会把被中途切断的一次性 child 归类为 `completed`。
+- 单元覆盖固定了无条件约定、每种终止原因、空闲与繁忙两种调度、批量语义、维护期回归、释放前顺序、父级已消失，以及一次不得让拆卸失败的发送被拒。
+- 三个 ACP 场景使用显式的结算栅栏，`subagent-report` 带有固定静默上报投递的配置 overlay。
+- 一个无密钥的 headless Loader 快照固定了「后台启动 → 管理器写入的结算通知 → 父级最终答案」路径，其中没有轮询，也没有 child `report` 调用。
+
+### 已接受的风险
+
+通知只是被投递，而不是被确认。没有持久化 mailbox、回执或重试：不在线的父级会丢失它，child 的 Session 仍是唯一的持久记录。要补上这一点，需要一套带有自身寻址、授权与重放规则的离线 mailbox 协议。
+
+当父级紧接着被 dispose 时（每个拆卸调用方都会这么做），在拆卸期间被 inject 的通知不会被模型读到：dispose 的 cancel 会清除这条未被认领的消息，而日志保留 insert/cancel 这一对作为记录。要让拆卸期投递在 resume 之后仍可读，要么需要上面那套离线 mailbox，要么需要改变 dispose 对持久待处理工作的处理方式。dispose 会丢弃每一条未被认领的 inbox 项，用户输入也不例外，因此改变该行为是一个 core-agent 决策，而不是结算投递的细节。resume 后的父级可以发现 child，但不会收到结局：`list_agents` 只报告存在性与「在线/仅存储」状态——`SubagentListEntry.activity` 就是这么写的——要取回结局，必须通过 `send_message` 去问那个 child。
+
+终止原因的归因是对日志既有 splice 词汇的尽力而为，偏向永不高估成功。`Inbox.remove()` 与拆卸的 `clear()` 写出的取消 splice 完全相同，因此删除一条内容仍保留在别处的消息——`workspace-context` 清理待处理的 instructions 刷新、或结算自身的 cancel 清掉一条仍在挂起的这类消息——可能被读作「工作被丢弃且从未运行」，把已完成的 child 报成被停下。区分二者需要 `dsh-agent` 提供更丰富的删除词汇；在该词汇可用前，这项误读的范围很窄，且错的方向是让父级复查一个已完成的 child，而永远不是信任一个未完成的 child。
+
+对于深或宽的树，轮次放大是真实存在的，而且按设计不可配置。step 边界的批量语义只能约束同时结算的情形，无法约束分散结算的 child。
+
+两个互相独立的唤醒源无法在手写 transcript 中排序。整体组装覆盖分别固定它们，而不固定它们的交错。
diff --git a/docs/config-catalog.i18n.yaml b/docs/config-catalog.i18n.yaml
index 0f6fc7c81d..476c7e69fe 100644
--- a/docs/config-catalog.i18n.yaml
+++ b/docs/config-catalog.i18n.yaml
@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write docs/config-catalog.md
-config-catalog.md: 2b57eefc30af4eb217cecbc255240468802b1f66
-config-catalog.zh.md: 51b332d0e60f8ff7dda49a95af377136cf6c432a
+config-catalog.md: 839f7b49c704727d972e507c43d932e59eb196b2
+config-catalog.zh.md: 8d26475b7ed0ceeab3e8c9ad6bfb639ea1c894dc
diff --git a/docs/config-catalog.md b/docs/config-catalog.md
index 2b57eefc30..839f7b49c7 100644
--- a/docs/config-catalog.md
+++ b/docs/config-catalog.md
@@ -2356,14 +2356,15 @@ Source: [`packages/subagent/tool-subagent/src/index.ts:25`](../packages/subagent
 
 ## `@deepseek-ai/dsh-tool-subagent-report`
 
-Requires: `subagents` · `tools`
+Requires: `subagents` · `tools` · `systemPrompt`
 
 ```ts config-catalog
 /** Config: how accepted reports are scheduled on the parent. */
 export interface Config {
   /**
-   * Parent scheduling (default `quiet`). `quiet` adds context without waking;
-   * `wakeup` creates one ordinary later parent turn.
+   * Parent scheduling (default `wakeup`). `wakeup` creates one ordinary later
+   * parent turn; `quiet` adds context without waking, so a parked parent learns
+   * of the report only when something else wakes it.
    */
   reportDelivery?: SubagentReportDelivery
 }
@@ -2371,7 +2372,7 @@ export interface Config {
 
 Depends on: [`SubagentReportDelivery`](subsystems/subagent.md)
 
-Source: [`packages/subagent/tool-subagent-report/src/index.ts:22`](../packages/subagent/tool-subagent-report/src/index.ts)
+Source: [`packages/subagent/tool-subagent-report/src/index.ts:27`](../packages/subagent/tool-subagent-report/src/index.ts)
 
 ## `@deepseek-ai/dsh-tool-tasks`
 
diff --git a/docs/config-catalog.zh.md b/docs/config-catalog.zh.md
index 51b332d0e6..8d26475b7e 100644
--- a/docs/config-catalog.zh.md
+++ b/docs/config-catalog.zh.md
@@ -2357,14 +2357,15 @@ export interface Config {
 
 ## `@deepseek-ai/dsh-tool-subagent-report`
 
-需要：`subagents` · `tools`
+需要：`subagents` · `tools` · `systemPrompt`
 
 ```ts config-catalog
 /** Config: how accepted reports are scheduled on the parent. */
 export interface Config {
   /**
-   * Parent scheduling (default `quiet`). `quiet` adds context without waking;
-   * `wakeup` creates one ordinary later parent turn.
+   * Parent scheduling (default `wakeup`). `wakeup` creates one ordinary later
+   * parent turn; `quiet` adds context without waking, so a parked parent learns
+   * of the report only when something else wakes it.
    */
   reportDelivery?: SubagentReportDelivery
 }
@@ -2372,7 +2373,7 @@ export interface Config {
 
 依赖：[`SubagentReportDelivery`](subsystems/subagent.md)
 
-来源：[`packages/subagent/tool-subagent-report/src/index.ts:22`](../packages/subagent/tool-subagent-report/src/index.ts)
+来源：[`packages/subagent/tool-subagent-report/src/index.ts:27`](../packages/subagent/tool-subagent-report/src/index.ts)
 
 ## `@deepseek-ai/dsh-tool-tasks`
 
diff --git a/docs/event-producer-consumer.i18n.yaml b/docs/event-producer-consumer.i18n.yaml
index 2d2341925b..2acd8e1b03 100644
--- a/docs/event-producer-consumer.i18n.yaml
+++ b/docs/event-producer-consumer.i18n.yaml
@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write docs/event-producer-consumer.md
-event-producer-consumer.md: 55a57480e0311aa047e9b5f0f90b6457fc9a007f
-event-producer-consumer.zh.md: c84c621befefdab7e668d64e90dcb14e28fd74ea
+event-producer-consumer.md: e14bad2b7a0f604f101281e271c4250a60b8c8eb
+event-producer-consumer.zh.md: 81ced8b25dc43a4f0d37b73ea954b1e4a997bdf9
diff --git a/docs/event-producer-consumer.md b/docs/event-producer-consumer.md
index 55a57480e0..e14bad2b7a 100644
--- a/docs/event-producer-consumer.md
+++ b/docs/event-producer-consumer.md
@@ -30,17 +30,17 @@ This matrix shows which packages dispatch each harness-owned event and which pac
 | `goal/changed` | `emit` | [`packages/goal/goal/src/domain.ts:114`](../packages/goal/goal/src/domain.ts) | [`goal`](../packages/goal/goal) (`emit`) | [`goal-session`](../packages/goal/goal-session) |
 | `llm/adapters-updated` | `emit` | [`packages/llm/llm/src/index.ts:75`](../packages/llm/llm/src/index.ts) | [`llm`](../packages/llm/llm) (`events.dispatch`) | `apiproxy`, [`llm`](../packages/llm/llm) |
 | `llm/stream` | `waterfall` | [`packages/llm/llm/src/index.ts:64`](../packages/llm/llm/src/index.ts) | [`llm`](../packages/llm/llm) (`waterfall`) | [`agent-loop`](../packages/core/agent-loop), [`llm`](../packages/llm/llm), [`llm-replay`](../packages/support/llm-replay), [`session-checkpoint-policy`](../packages/session/session-checkpoint-policy), [`session-title`](../packages/session/session-title) |
-| `session/created` | `emit` | [`packages/core/session/src/index.ts:75`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | `apiproxy`, [`compact`](../packages/compact/compact), [`goal`](../packages/goal/goal), [`hook-protocol`](../packages/hooks/hook-protocol), [`llm-retry`](../packages/llm/llm-retry), [`permission`](../packages/interaction/permission), [`plan-mode`](../packages/plan/plan-mode), `server`, [`session`](../packages/core/session), [`session-persistence`](../packages/session/session-persistence), [`session-telemetry`](../packages/session/session-telemetry), [`tools`](../packages/core/tools), [`user-approval`](../packages/interaction/user-approval) |
-| `session/disposed` | `emit` | [`packages/core/session/src/index.ts:85`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | [`agent-loop`](../packages/core/agent-loop), `apiproxy`, [`session-persistence`](../packages/session/session-persistence), [`session-projection-cache`](../packages/session/session-projection-cache), [`session-telemetry`](../packages/session/session-telemetry), [`session-title`](../packages/session/session-title) |
-| `session/event` | `emit` | [`packages/core/session/src/index.ts:97`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | [`acp`](../packages/acp/acp), [`agent-loop`](../packages/core/agent-loop), `apiproxy`, [`compact`](../packages/compact/compact), [`compact-basic`](../packages/compact/compact-basic), [`goal`](../packages/goal/goal), [`goal-session`](../packages/goal/goal-session), [`hook-protocol`](../packages/hooks/hook-protocol), [`loader-smoke`](../packages/support/loader-smoke), `server`, [`session`](../packages/core/session), [`session-persistence`](../packages/session/session-persistence), [`session-projection`](../packages/session/session-projection), [`session-projection-cache`](../packages/session/session-projection-cache), [`session-telemetry`](../packages/session/session-telemetry), [`session-telemetry-otel`](../packages/session/session-telemetry-otel), [`session-title`](../packages/session/session-title), [`token-meter`](../packages/llm/token-meter), [`tools`](../packages/core/tools), [`user-approval`](../packages/interaction/user-approval), [`workspace-context`](../packages/context/workspace-context) |
-| `session/flush` | `parallel` | [`packages/core/session/src/index.ts:106`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | [`session-persistence`](../packages/session/session-persistence), [`session-telemetry`](../packages/session/session-telemetry) |
+| `session/created` | `emit` | [`packages/core/session/src/index.ts:54`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | `apiproxy`, [`compact`](../packages/compact/compact), [`goal`](../packages/goal/goal), [`hook-protocol`](../packages/hooks/hook-protocol), [`llm-retry`](../packages/llm/llm-retry), [`permission`](../packages/interaction/permission), [`plan-mode`](../packages/plan/plan-mode), `server`, [`session`](../packages/core/session), [`session-persistence`](../packages/session/session-persistence), [`session-telemetry`](../packages/session/session-telemetry), [`tools`](../packages/core/tools), [`user-approval`](../packages/interaction/user-approval) |
+| `session/disposed` | `emit` | [`packages/core/session/src/index.ts:64`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | [`agent-loop`](../packages/core/agent-loop), `apiproxy`, [`session-persistence`](../packages/session/session-persistence), [`session-projection-cache`](../packages/session/session-projection-cache), [`session-telemetry`](../packages/session/session-telemetry), [`session-title`](../packages/session/session-title) |
+| `session/event` | `emit` | [`packages/core/session/src/index.ts:76`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | [`acp`](../packages/acp/acp), [`agent-loop`](../packages/core/agent-loop), `apiproxy`, [`compact`](../packages/compact/compact), [`compact-basic`](../packages/compact/compact-basic), [`goal`](../packages/goal/goal), [`goal-session`](../packages/goal/goal-session), [`hook-protocol`](../packages/hooks/hook-protocol), [`loader-smoke`](../packages/support/loader-smoke), `server`, [`session`](../packages/core/session), [`session-persistence`](../packages/session/session-persistence), [`session-projection`](../packages/session/session-projection), [`session-projection-cache`](../packages/session/session-projection-cache), [`session-telemetry`](../packages/session/session-telemetry), [`session-telemetry-otel`](../packages/session/session-telemetry-otel), [`session-title`](../packages/session/session-title), [`token-meter`](../packages/llm/token-meter), [`tools`](../packages/core/tools), [`user-approval`](../packages/interaction/user-approval), [`workspace-context`](../packages/context/workspace-context) |
+| `session/flush` | `parallel` | [`packages/core/session/src/index.ts:85`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | [`session-persistence`](../packages/session/session-persistence), [`session-telemetry`](../packages/session/session-telemetry) |
 | `settings/document-updated` | `emit` | [`packages/settings/settings/src/index.ts:170`](../packages/settings/settings/src/index.ts) | [`settings`](../packages/settings/settings) (`events.dispatch`) | `apiproxy` |
 | `settings/updated` | `emit` | [`packages/settings/settings/src/index.ts:157`](../packages/settings/settings/src/index.ts) | [`settings`](../packages/settings/settings) (`events.dispatch`) | [`settings`](../packages/settings/settings) |
 | `skills/change` | `emit` | [`packages/skill/skill/src/index.ts:297`](../packages/skill/skill/src/index.ts) | [`skill`](../packages/skill/skill) (`events.dispatch`) | - |
-| `subagent/end` | `emit` | [`packages/subagent/subagent/src/index.ts:165`](../packages/subagent/subagent/src/index.ts) | [`subagent`](../packages/subagent/subagent) (`events.dispatch`) | [`hooks-claude`](../packages/hooks/hooks-claude), `server`, [`subagent`](../packages/subagent/subagent) |
-| `subagent/provider-added` | `emit` | [`packages/subagent/subagent/src/index.ts:139`](../packages/subagent/subagent/src/index.ts) | [`subagent`](../packages/subagent/subagent) (`emit`) | [`subagent`](../packages/subagent/subagent), [`tool-subagent`](../packages/subagent/tool-subagent) |
-| `subagent/provider-removed` | `emit` | [`packages/subagent/subagent/src/index.ts:145`](../packages/subagent/subagent/src/index.ts) | [`subagent`](../packages/subagent/subagent) (`events.dispatch`) | [`subagent`](../packages/subagent/subagent), [`tool-subagent`](../packages/subagent/tool-subagent) |
-| `subagent/start` | `emit` | [`packages/subagent/subagent/src/index.ts:156`](../packages/subagent/subagent/src/index.ts) | [`subagent`](../packages/subagent/subagent) (`events.dispatch`) | [`hooks-claude`](../packages/hooks/hooks-claude), [`subagent`](../packages/subagent/subagent) |
+| `subagent/end` | `emit` | [`packages/subagent/subagent/src/index.ts:166`](../packages/subagent/subagent/src/index.ts) | [`subagent`](../packages/subagent/subagent) (`events.dispatch`) | [`hooks-claude`](../packages/hooks/hooks-claude), `server`, [`subagent`](../packages/subagent/subagent) |
+| `subagent/provider-added` | `emit` | [`packages/subagent/subagent/src/index.ts:140`](../packages/subagent/subagent/src/index.ts) | [`subagent`](../packages/subagent/subagent) (`emit`) | [`subagent`](../packages/subagent/subagent), [`tool-subagent`](../packages/subagent/tool-subagent) |
+| `subagent/provider-removed` | `emit` | [`packages/subagent/subagent/src/index.ts:146`](../packages/subagent/subagent/src/index.ts) | [`subagent`](../packages/subagent/subagent) (`events.dispatch`) | [`subagent`](../packages/subagent/subagent), [`tool-subagent`](../packages/subagent/tool-subagent) |
+| `subagent/start` | `emit` | [`packages/subagent/subagent/src/index.ts:157`](../packages/subagent/subagent/src/index.ts) | [`subagent`](../packages/subagent/subagent) (`events.dispatch`) | [`hooks-claude`](../packages/hooks/hooks-claude), [`subagent`](../packages/subagent/subagent) |
 | `system-prompt/assemble` | `waterfall` | [`packages/core/system-prompt/src/index.ts:31`](../packages/core/system-prompt/src/index.ts) | [`system-prompt`](../packages/core/system-prompt) (`waterfall`) | [`agent`](../packages/core/agent), [`agent-presets`](../packages/preset/agent-presets), [`system-prompt`](../packages/core/system-prompt) |
 | `system-prompt/change` | `emit` | [`packages/core/system-prompt/src/index.ts:37`](../packages/core/system-prompt/src/index.ts) | [`system-prompt`](../packages/core/system-prompt) (`emit`) | - |
 | `telemetry/record` | `waterfall` | [`packages/session/session-telemetry/src/index.ts:43`](../packages/session/session-telemetry/src/index.ts) | [`session-telemetry`](../packages/session/session-telemetry) (`waterfall`) | - |
diff --git a/docs/event-producer-consumer.zh.md b/docs/event-producer-consumer.zh.md
index c84c621bef..81ced8b25d 100644
--- a/docs/event-producer-consumer.zh.md
+++ b/docs/event-producer-consumer.zh.md
@@ -32,10 +32,10 @@
 | `goal/changed` | `emit` | [`packages/goal/goal/src/domain.ts:114`](../packages/goal/goal/src/domain.ts) | [`goal`](../packages/goal/goal) (`emit`) | [`goal-session`](../packages/goal/goal-session) |
 | `llm/adapters-updated` | `emit` | [`packages/llm/llm/src/index.ts:75`](../packages/llm/llm/src/index.ts) | [`llm`](../packages/llm/llm) (`events.dispatch`) | `apiproxy`, [`llm`](../packages/llm/llm) |
 | `llm/stream` | `waterfall` | [`packages/llm/llm/src/index.ts:64`](../packages/llm/llm/src/index.ts) | [`llm`](../packages/llm/llm) (`waterfall`) | [`agent-loop`](../packages/core/agent-loop), [`llm`](../packages/llm/llm), [`llm-replay`](../packages/support/llm-replay), [`session-checkpoint-policy`](../packages/session/session-checkpoint-policy), [`session-title`](../packages/session/session-title) |
-| `session/created` | `emit` | [`packages/core/session/src/index.ts:75`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | `apiproxy`, [`compact`](../packages/compact/compact), [`goal`](../packages/goal/goal), [`hook-protocol`](../packages/hooks/hook-protocol), [`llm-retry`](../packages/llm/llm-retry), [`permission`](../packages/interaction/permission), [`plan-mode`](../packages/plan/plan-mode), `server`, [`session`](../packages/core/session), [`session-persistence`](../packages/session/session-persistence), [`session-telemetry`](../packages/session/session-telemetry), [`tools`](../packages/core/tools), [`user-approval`](../packages/interaction/user-approval) |
-| `session/disposed` | `emit` | [`packages/core/session/src/index.ts:85`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | [`agent-loop`](../packages/core/agent-loop), `apiproxy`, [`session-persistence`](../packages/session/session-persistence), [`session-projection-cache`](../packages/session/session-projection-cache), [`session-telemetry`](../packages/session/session-telemetry), [`session-title`](../packages/session/session-title) |
-| `session/event` | `emit` | [`packages/core/session/src/index.ts:97`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | [`acp`](../packages/acp/acp), [`agent-loop`](../packages/core/agent-loop), `apiproxy`, [`compact`](../packages/compact/compact), [`compact-basic`](../packages/compact/compact-basic), [`goal`](../packages/goal/goal), [`goal-session`](../packages/goal/goal-session), [`hook-protocol`](../packages/hooks/hook-protocol), [`loader-smoke`](../packages/support/loader-smoke), `server`, [`session`](../packages/core/session), [`session-persistence`](../packages/session/session-persistence), [`session-projection`](../packages/session/session-projection), [`session-projection-cache`](../packages/session/session-projection-cache), [`session-telemetry`](../packages/session/session-telemetry), [`session-telemetry-otel`](../packages/session/session-telemetry-otel), [`session-title`](../packages/session/session-title), [`token-meter`](../packages/llm/token-meter), [`tools`](../packages/core/tools), [`user-approval`](../packages/interaction/user-approval), [`workspace-context`](../packages/context/workspace-context) |
-| `session/flush` | `parallel` | [`packages/core/session/src/index.ts:106`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | [`session-persistence`](../packages/session/session-persistence), [`session-telemetry`](../packages/session/session-telemetry) |
+| `session/created` | `emit` | [`packages/core/session/src/index.ts:54`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | `apiproxy`, [`compact`](../packages/compact/compact), [`goal`](../packages/goal/goal), [`hook-protocol`](../packages/hooks/hook-protocol), [`llm-retry`](../packages/llm/llm-retry), [`permission`](../packages/interaction/permission), [`plan-mode`](../packages/plan/plan-mode), `server`, [`session`](../packages/core/session), [`session-persistence`](../packages/session/session-persistence), [`session-telemetry`](../packages/session/session-telemetry), [`tools`](../packages/core/tools), [`user-approval`](../packages/interaction/user-approval) |
+| `session/disposed` | `emit` | [`packages/core/session/src/index.ts:64`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | [`agent-loop`](../packages/core/agent-loop), `apiproxy`, [`session-persistence`](../packages/session/session-persistence), [`session-projection-cache`](../packages/session/session-projection-cache), [`session-telemetry`](../packages/session/session-telemetry), [`session-title`](../packages/session/session-title) |
+| `session/event` | `emit` | [`packages/core/session/src/index.ts:76`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | [`acp`](../packages/acp/acp), [`agent-loop`](../packages/core/agent-loop), `apiproxy`, [`compact`](../packages/compact/compact), [`compact-basic`](../packages/compact/compact-basic), [`goal`](../packages/goal/goal), [`goal-session`](../packages/goal/goal-session), [`hook-protocol`](../packages/hooks/hook-protocol), [`loader-smoke`](../packages/support/loader-smoke), `server`, [`session`](../packages/core/session), [`session-persistence`](../packages/session/session-persistence), [`session-projection`](../packages/session/session-projection), [`session-projection-cache`](../packages/session/session-projection-cache), [`session-telemetry`](../packages/session/session-telemetry), [`session-telemetry-otel`](../packages/session/session-telemetry-otel), [`session-title`](../packages/session/session-title), [`token-meter`](../packages/llm/token-meter), [`tools`](../packages/core/tools), [`user-approval`](../packages/interaction/user-approval), [`workspace-context`](../packages/context/workspace-context) |
+| `session/flush` | `parallel` | [`packages/core/session/src/index.ts:85`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | [`session-persistence`](../packages/session/session-persistence), [`session-telemetry`](../packages/session/session-telemetry) |
 | `settings/document-updated` | `emit` | [`packages/settings/settings/src/index.ts:170`](../packages/settings/settings/src/index.ts) | [`settings`](../packages/settings/settings) (`events.dispatch`) | `apiproxy` |
 | `settings/updated` | `emit` | [`packages/settings/settings/src/index.ts:157`](../packages/settings/settings/src/index.ts) | [`settings`](../packages/settings/settings) (`events.dispatch`) | [`settings`](../packages/settings/settings) |
 | `skills/change` | `emit` | [`packages/skill/skill/src/index.ts:297`](../packages/skill/skill/src/index.ts) | [`skill`](../packages/skill/skill) (`events.dispatch`) | - |
diff --git a/docs/module-graph.i18n.yaml b/docs/module-graph.i18n.yaml
index 964e51e8d1..dd4924f958 100644
--- a/docs/module-graph.i18n.yaml
+++ b/docs/module-graph.i18n.yaml
@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write docs/module-graph.md
-module-graph.md: 8ee17f68d25d7ae693955eae6baa70675347ad71
-module-graph.zh.md: 9f89a5b8b58d55ad9ad4675ebe702140f8253b42
+module-graph.md: 7e05f21a6d153bc91880721bda0821a92ea6da37
+module-graph.zh.md: 1eef008ddba5b290039f99dae408adde19ede5ee
diff --git a/docs/module-graph.md b/docs/module-graph.md
index 8ee17f68d2..7e05f21a6d 100644
--- a/docs/module-graph.md
+++ b/docs/module-graph.md
@@ -1047,6 +1047,7 @@ flowchart TD
   pkg_tool_subagent_report --> pkg_invariants
   pkg_tool_subagent_report --> pkg_llm
   pkg_tool_subagent_report --> pkg_subagent
+  pkg_tool_subagent_report --> pkg_system_prompt
   pkg_tool_subagent_report --> pkg_tools
   pkg_hooks_claude --> pkg_agent
   pkg_hooks_claude --> pkg_hook_protocol
@@ -1435,7 +1436,7 @@ flowchart TD
 | [`subagent-inprocess`](../packages/subagent/subagent-inprocess) | `subagent` | [`agent`](../packages/core/agent), [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`session`](../packages/core/session), [`subagent`](../packages/subagent/subagent), [`system-prompt`](../packages/core/system-prompt), [`tools`](../packages/core/tools) |
 | [`tool-subagent`](../packages/subagent/tool-subagent) | `subagent` | [`agent`](../packages/core/agent), [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`subagent`](../packages/subagent/subagent), [`tasks`](../packages/tasks/tasks), [`tools`](../packages/core/tools) |
 | [`tool-subagent-control`](../packages/subagent/tool-subagent-control) | `subagent` | [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`session`](../packages/core/session), [`subagent`](../packages/subagent/subagent), [`tools`](../packages/core/tools) |
-| [`tool-subagent-report`](../packages/subagent/tool-subagent-report) | `subagent` | [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`subagent`](../packages/subagent/subagent), [`tools`](../packages/core/tools) |
+| [`tool-subagent-report`](../packages/subagent/tool-subagent-report) | `subagent` | [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`subagent`](../packages/subagent/subagent), [`system-prompt`](../packages/core/system-prompt), [`tools`](../packages/core/tools) |
 | [`hooks-claude`](../packages/hooks/hooks-claude) | `hooks` | [`agent`](../packages/core/agent), [`hook-protocol`](../packages/hooks/hook-protocol), [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`session`](../packages/core/session), [`session-persistence`](../packages/session/session-persistence), [`subagent`](../packages/subagent/subagent), [`tools`](../packages/core/tools) |
 | [`web-app`](../packages/bundle/web-app) | `bundle` | [`bash-env`](../packages/bash/bash-env), [`invariants`](../packages/support/invariants), [`system-prompt`](../packages/core/system-prompt) |
 | [`client-ui-conversation`](../packages/client/ui-conversation) | `client` | [`agent`](../packages/core/agent), [`attachment`](../packages/attachment/attachment), [`brand`](../packages/util/brand), [`client-connection`](../packages/client/connection), [`client-locale`](../packages/client/locale), [`client-runtime`](../packages/client/runtime), [`client-ui-primitives`](../packages/client/ui-primitives), [`client-ui-slash`](../packages/client/ui-slash), [`client-ui-slots`](../packages/client/ui-slots), [`commands`](../packages/interaction/commands), [`compact`](../packages/compact/compact), [`invariants`](../packages/support/invariants), [`llm-retry`](../packages/llm/llm-retry), [`token-meter`](../packages/llm/token-meter), [`tools`](../packages/core/tools) |
diff --git a/docs/module-graph.zh.md b/docs/module-graph.zh.md
index 9f89a5b8b5..1eef008ddb 100644
--- a/docs/module-graph.zh.md
+++ b/docs/module-graph.zh.md
@@ -1049,6 +1049,7 @@ flowchart TD
   pkg_tool_subagent_report --> pkg_invariants
   pkg_tool_subagent_report --> pkg_llm
   pkg_tool_subagent_report --> pkg_subagent
+  pkg_tool_subagent_report --> pkg_system_prompt
   pkg_tool_subagent_report --> pkg_tools
   pkg_hooks_claude --> pkg_agent
   pkg_hooks_claude --> pkg_hook_protocol
@@ -1437,7 +1438,7 @@ flowchart TD
 | [`subagent-inprocess`](../packages/subagent/subagent-inprocess) | `subagent` | [`agent`](../packages/core/agent), [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`session`](../packages/core/session), [`subagent`](../packages/subagent/subagent), [`system-prompt`](../packages/core/system-prompt), [`tools`](../packages/core/tools) |
 | [`tool-subagent`](../packages/subagent/tool-subagent) | `subagent` | [`agent`](../packages/core/agent), [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`subagent`](../packages/subagent/subagent), [`tasks`](../packages/tasks/tasks), [`tools`](../packages/core/tools) |
 | [`tool-subagent-control`](../packages/subagent/tool-subagent-control) | `subagent` | [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`session`](../packages/core/session), [`subagent`](../packages/subagent/subagent), [`tools`](../packages/core/tools) |
-| [`tool-subagent-report`](../packages/subagent/tool-subagent-report) | `subagent` | [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`subagent`](../packages/subagent/subagent), [`tools`](../packages/core/tools) |
+| [`tool-subagent-report`](../packages/subagent/tool-subagent-report) | `subagent` | [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`subagent`](../packages/subagent/subagent), [`system-prompt`](../packages/core/system-prompt), [`tools`](../packages/core/tools) |
 | [`hooks-claude`](../packages/hooks/hooks-claude) | `hooks` | [`agent`](../packages/core/agent), [`hook-protocol`](../packages/hooks/hook-protocol), [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`session`](../packages/core/session), [`session-persistence`](../packages/session/session-persistence), [`subagent`](../packages/subagent/subagent), [`tools`](../packages/core/tools) |
 | [`web-app`](../packages/bundle/web-app) | `bundle` | [`bash-env`](../packages/bash/bash-env), [`invariants`](../packages/support/invariants), [`system-prompt`](../packages/core/system-prompt) |
 | [`client-ui-conversation`](../packages/client/ui-conversation) | `client` | [`agent`](../packages/core/agent), [`attachment`](../packages/attachment/attachment), [`brand`](../packages/util/brand), [`client-connection`](../packages/client/connection), [`client-locale`](../packages/client/locale), [`client-runtime`](../packages/client/runtime), [`client-ui-primitives`](../packages/client/ui-primitives), [`client-ui-slash`](../packages/client/ui-slash), [`client-ui-slots`](../packages/client/ui-slots), [`commands`](../packages/interaction/commands), [`compact`](../packages/compact/compact), [`invariants`](../packages/support/invariants), [`llm-retry`](../packages/llm/llm-retry), [`token-meter`](../packages/llm/token-meter), [`tools`](../packages/core/tools) |
diff --git a/docs/subsystems/core.i18n.yaml b/docs/subsystems/core.i18n.yaml
index 0123946525..cb30300ba5 100644
--- a/docs/subsystems/core.i18n.yaml
+++ b/docs/subsystems/core.i18n.yaml
@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write docs/subsystems/core.md
-core.md: 96655026f5affda6fed080496d975e2366f0356f
-core.zh.md: e2cde8845ddf6b78f64d062fd8860c0c88b7ce11
+core.md: c9d8eb1b854432668092754206535a6985ebc6fa
+core.zh.md: 73f9954959df692f91f43bc7bf8e02bbfa77c6b3
diff --git a/docs/subsystems/core.md b/docs/subsystems/core.md
index 96655026f5..c9d8eb1b85 100644
--- a/docs/subsystems/core.md
+++ b/docs/subsystems/core.md
@@ -718,7 +718,7 @@ list(): Agent[]
 roots(): Agent[]
 ```
 
-Source: [`packages/core/agent/src/index.ts:255`](../../packages/core/agent/src/index.ts)
+Source: [`packages/core/agent/src/index.ts:256`](../../packages/core/agent/src/index.ts)
 
 <a id="agent-events"></a>
 
diff --git a/docs/subsystems/core.zh.md b/docs/subsystems/core.zh.md
index e2cde8845d..73f9954959 100644
--- a/docs/subsystems/core.zh.md
+++ b/docs/subsystems/core.zh.md
@@ -726,7 +726,7 @@ list(): Agent[]
 roots(): Agent[]
 ```
 
-Source: [`packages/core/agent/src/index.ts:255`](../../packages/core/agent/src/index.ts)
+Source: [`packages/core/agent/src/index.ts:256`](../../packages/core/agent/src/index.ts)
 
 <a id="agent-events"></a>
 
diff --git a/docs/subsystems/session.i18n.yaml b/docs/subsystems/session.i18n.yaml
index 177e33013d..37eeff3722 100644
--- a/docs/subsystems/session.i18n.yaml
+++ b/docs/subsystems/session.i18n.yaml
@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write docs/subsystems/session.md
-session.md: 990b249cde9f02343f2c668aee5d7c000837df56
-session.zh.md: 39e8ff1e8831fd75c8929c93e263622bb5aa6ea4
+session.md: e81f284f54057aa92c5c8cc76c2200751e51f4d7
+session.zh.md: 2d9eb45c4addf7c8df2370f9a0f3b4b57b4f05a1
diff --git a/docs/subsystems/session.md b/docs/subsystems/session.md
index 990b249cde..e81f284f54 100644
--- a/docs/subsystems/session.md
+++ b/docs/subsystems/session.md
@@ -744,7 +744,7 @@ fork(source: SessionForkSource, boundary?: number, childSessionId?: SessionId):
 
 Types: [CreateSessionOptions](persistence.md) · [PrepareSessionOptions](persistence.md) · [SessionId](core.md)
 
-Source: [`packages/core/session/src/index.ts:813`](../../packages/core/session/src/index.ts)
+Source: [`packages/core/session/src/index.ts:792`](../../packages/core/session/src/index.ts)
 
 <a id="session-events"></a>
 
@@ -773,7 +773,7 @@ Creation announcement during session publication. A synchronous throw vetoes and
 
 Types: [Scoped](scope.md)
 
-Source: [`packages/core/session/src/index.ts:75`](../../packages/core/session/src/index.ts)
+Source: [`packages/core/session/src/index.ts:54`](../../packages/core/session/src/index.ts)
 
 <a id="sessiondisposed--emit"></a>
 
@@ -796,7 +796,7 @@ Emitted once when an announced session leaves the store, including publication r
 
 Types: [Scoped](scope.md)
 
-Source: [`packages/core/session/src/index.ts:85`](../../packages/core/session/src/index.ts)
+Source: [`packages/core/session/src/index.ts:64`](../../packages/core/session/src/index.ts)
 
 <a id="sessionevent--emit"></a>
 
@@ -821,7 +821,7 @@ Post-commit, fire-and-forget append feed. The listener snapshot resolves before
 
 Types: [Scoped](scope.md)
 
-Source: [`packages/core/session/src/index.ts:97`](../../packages/core/session/src/index.ts)
+Source: [`packages/core/session/src/index.ts:76`](../../packages/core/session/src/index.ts)
 
 <a id="sessionflush--parallel"></a>
 
@@ -843,5 +843,5 @@ Awaited parallel durability checkpoint: every listener runs and the caller await
 
 Types: [Scoped](scope.md)
 
-Source: [`packages/core/session/src/index.ts:106`](../../packages/core/session/src/index.ts)
+Source: [`packages/core/session/src/index.ts:85`](../../packages/core/session/src/index.ts)
 <!-- END GENERATED cordis-surface -->
diff --git a/docs/subsystems/session.zh.md b/docs/subsystems/session.zh.md
index 39e8ff1e88..2d9eb45c4a 100644
--- a/docs/subsystems/session.zh.md
+++ b/docs/subsystems/session.zh.md
@@ -748,7 +748,7 @@ fork(source: SessionForkSource, boundary?: number, childSessionId?: SessionId):
 
 Types: [CreateSessionOptions](persistence.md) · [PrepareSessionOptions](persistence.md) · [SessionId](core.md)
 
-Source: [`packages/core/session/src/index.ts:813`](../../packages/core/session/src/index.ts)
+Source: [`packages/core/session/src/index.ts:792`](../../packages/core/session/src/index.ts)
 
 <a id="session-events"></a>
 
@@ -777,7 +777,7 @@ Creation announcement during session publication. A synchronous throw vetoes and
 
 Types: [Scoped](scope.md)
 
-Source: [`packages/core/session/src/index.ts:75`](../../packages/core/session/src/index.ts)
+Source: [`packages/core/session/src/index.ts:54`](../../packages/core/session/src/index.ts)
 
 <a id="sessiondisposed--emit"></a>
 
@@ -800,7 +800,7 @@ Emitted once when an announced session leaves the store, including publication r
 
 Types: [Scoped](scope.md)
 
-Source: [`packages/core/session/src/index.ts:85`](../../packages/core/session/src/index.ts)
+Source: [`packages/core/session/src/index.ts:64`](../../packages/core/session/src/index.ts)
 
 <a id="sessionevent--emit"></a>
 
@@ -825,7 +825,7 @@ Post-commit, fire-and-forget append feed. The listener snapshot resolves before
 
 Types: [Scoped](scope.md)
 
-Source: [`packages/core/session/src/index.ts:97`](../../packages/core/session/src/index.ts)
+Source: [`packages/core/session/src/index.ts:76`](../../packages/core/session/src/index.ts)
 
 <a id="sessionflush--parallel"></a>
 
@@ -847,5 +847,5 @@ Awaited parallel durability checkpoint: every listener runs and the caller await
 
 Types: [Scoped](scope.md)
 
-Source: [`packages/core/session/src/index.ts:106`](../../packages/core/session/src/index.ts)
+Source: [`packages/core/session/src/index.ts:85`](../../packages/core/session/src/index.ts)
 <!-- END GENERATED cordis-surface -->
diff --git a/docs/subsystems/subagent.i18n.yaml b/docs/subsystems/subagent.i18n.yaml
index a5767904b9..7c195c00f0 100644
--- a/docs/subsystems/subagent.i18n.yaml
+++ b/docs/subsystems/subagent.i18n.yaml
@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write docs/subsystems/subagent.md
-subagent.md: cf728d9f15fdaaf8199954e91b564167e8efe439
-subagent.zh.md: 1a8e2e5837b8ac88f7d7b7ca767fb7aa21391a9f
+subagent.md: 71ad9bbb6d3093aefd6c0ebcf891330cc0ed707e
+subagent.zh.md: 13754acc0606354e823abe7ba1d17ec295ec6670
diff --git a/docs/subsystems/subagent.md b/docs/subsystems/subagent.md
index cf728d9f15..71ad9bbb6d 100644
--- a/docs/subsystems/subagent.md
+++ b/docs/subsystems/subagent.md
@@ -209,6 +209,27 @@ interface SubagentReportMessageSource {
 type SubagentReportDelivery = 'quiet' | 'wakeup'
 ```
 
+Reporting is the child's own choice, so the manager keeps a separate account of its own: when a resident Activation settles, it delivers one notice to the child's durable direct parent describing how that epoch ended and carrying its final assistant content. That delivery is unconditional for every child whose id a caller received, happens before the ownership release that would let the parent be judged settled, and reaches a resident parent through the same waking-admission accounting as a report. A parent whose own lineage is already tearing down receives it without a wake, because waking a quiescent Agent starts a turn rather than queueing work. Its provenance is a distinct kind so a transcript never presents a runtime account as something the child wrote.
+
+```ts type-equiv
+/**
+ * Durable attribution for the runtime's own account of a continuable child
+ * settling. Deliberately a different kind from
+ * {@link SubagentReportMessageSource}: a report is content the child chose,
+ * while this message is the manager stating what became of the child, and a
+ * transcript that merged them would credit the child with words it never wrote.
+ */
+interface SubagentSettledMessageSource {
+  readonly kind: 'subagent-settled'
+  /** A runtime account shown without expanding the row (`notice` context form). */
+  readonly form: 'notice'
+  /** One-line account of how the child ended. */
+  readonly summary: string
+  /** Session id of the child that settled. */
+  readonly senderSessionId: SessionId
+}
+```
+
 ```ts type-equiv
 /** Options for one continuable child's report to its direct parent. */
 interface SubagentReportOptions {
@@ -265,7 +286,7 @@ A local one-shot provider appends the descriptor inside the child's initial turn
 
 ## Durable enumeration: `listChildren()`, `listDescendants()`, and their entries
 
-`SubagentService.listChildren(parentSessionId)` enumerates the parent's direct session-backed subagents from the live-preferred merge of `ctx.sessions.list()` and optional `ctx.sessionPersistence.list()` — no query service, and no Agent is loaded or resumed. Candidates are the direct children whose durable header carries `origin: 'subagent'`; the marker classifies enumeration and coarse generic-route denial but cannot establish a valid descriptor, resumability, or authorization — the projection fold owns identity, and the Activation contract owns resume. Each row's `mode`/`label` is the registered `subagent` projection unit's value, served through a three-rung ladder: the registry's watermark cache for a live child (zero log reads); the optional projection checkpoint cache for a cold one (`cachedSnapshot` — an identity passing the own-suffix seq gate is final, because an own descriptor is immutable once appended); otherwise one `persistence.inspect()` reading folded through the registry (bounded concurrency, recomputed per listing). The cache is a pure optional accelerator: absent, serving the `null` sentinel or missing the key, failing the seq gate, or faulting, it falls silently through to the authoritative refold. The fold is `subagent/descriptor` last-wins with no failure channel: the child's own descriptor overrides a fork-seeded ancestor's, and a malformed or unknown-version payload folds to a serializable `null` sentinel, treated as no value. The result is one `SubagentListEntry[]` in `createdAt`-then-id order: a served identity yields a `child` entry with `mode: 'one-shot' | 'continuable'` and `activity: 'running' | 'inactive'`; continuable entries always carry `label`, while one-shot entries carry it only when the start caller supplied presentation metadata. A settled candidate whose fold served no identity yields a `corrupt` diagnostic — missing, malformed, and unknown-version descriptors deliberately undistinguished (`unsupported` remains in the type but is never produced); a running candidate without an identity is omitted (the creation window before its descriptor lands); a failed cold inspection yields one `unavailable` diagnostic retried on the next listing, so one damaged sibling cannot hide healthy children. `hasChildren` marks a direct descendant with durable subagent origin, read from the same merged material. Activity snapshots only whether the logical record is live in `ctx.sessions`, not outcome or resumability. Absent persistence, enumeration is live-only rather than an error — a cold child cannot be resumed then either. `listChildren()` throws `SubagentError` with code `SUBAGENT_CONTROL_PROJECTIONS_UNAVAILABLE` when the `ctx.sessionProjections` registry is absent and `SUBAGENT_CONTROL_SESSION_STORE_UNAVAILABLE` when the session store is, both checked before any read so a deployment with zero children still fails deterministically; the list tool requires `ctx.subagents` and `ctx.agents` at plugin load. A service consumer such as a UI can display both modes and choose an unlabeled one-shot fallback, while the model-facing `list_agents` adapter (the separately loadable `/list-agents` plugin of [dsh-tool-subagent-control](../../packages/subagent/tool-subagent-control)) keeps only continuable entries and refines status through the live Agent registry's `running`/`idle`/`complete` vocabulary. Listing does not consult the continuation manager's Activation map, Agent registry, or provider availability; `send_message` remains the authoritative delivery-time operation, and a listed running continuable child may still reject delivery as an ownership conflict. The read-path rationale lives in [the list-identity-projection Agent Note](../../.agents/notes/implemented/architecture/2026-08-06-subagent-list-identity-projection.md).
+`SubagentService.listChildren(parentSessionId)` enumerates the parent's direct session-backed subagents from the live-preferred merge of `ctx.sessions.list()` and optional `ctx.sessionPersistence.list()` — no query service, and no Agent is loaded or resumed. Candidates are the direct children whose durable header carries `origin: 'subagent'`; the marker classifies enumeration and coarse generic-route denial but cannot establish a valid descriptor, resumability, or authorization — the projection fold owns identity, and the Activation contract owns resume. Each row's `mode`/`label` is the registered `subagent` projection unit's value, served through a three-rung ladder: the registry's watermark cache for a live child (zero log reads); the optional projection checkpoint cache for a cold one (`cachedSnapshot` — an identity passing the own-suffix seq gate is final, because an own descriptor is immutable once appended); otherwise one `persistence.inspect()` reading folded through the registry (bounded concurrency, recomputed per listing). The cache is a pure optional accelerator: absent, serving the `null` sentinel or missing the key, failing the seq gate, or faulting, it falls silently through to the authoritative refold. The fold is `subagent/descriptor` last-wins with no failure channel: the child's own descriptor overrides a fork-seeded ancestor's, and a malformed or unknown-version payload folds to a serializable `null` sentinel, treated as no value. The result is one `SubagentListEntry[]` in `createdAt`-then-id order: a served identity yields a `child` entry with `mode: 'one-shot' | 'continuable'` and `activity: 'running' | 'inactive'`; continuable entries always carry `label`, while one-shot entries carry it only when the start caller supplied presentation metadata. A settled candidate whose fold served no identity yields a `corrupt` diagnostic — missing, malformed, and unknown-version descriptors deliberately undistinguished (`unsupported` remains in the type but is never produced); a running candidate without an identity is omitted (the creation window before its descriptor lands); a failed cold inspection yields one `unavailable` diagnostic retried on the next listing, so one damaged sibling cannot hide healthy children. `hasChildren` marks a direct descendant with durable subagent origin, read from the same merged material. Activity snapshots only whether the logical record is live in `ctx.sessions`, not outcome or resumability. Absent persistence, enumeration is live-only rather than an error — a cold child cannot be resumed then either. `listChildren()` throws `SubagentError` with code `SUBAGENT_CONTROL_PROJECTIONS_UNAVAILABLE` when the `ctx.sessionProjections` registry is absent and `SUBAGENT_CONTROL_SESSION_STORE_UNAVAILABLE` when the session store is, both checked before any read so a deployment with zero children still fails deterministically; the list tool requires `ctx.subagents` and `ctx.agents` at plugin load. A service consumer such as a UI can display both modes and choose an unlabeled one-shot fallback, while the model-facing `list_agents` adapter (the separately loadable `/list-agents` plugin of [dsh-tool-subagent-control](../../packages/subagent/tool-subagent-control)) keeps only continuable entries and refines status through the live Agent registry into its own `running`/`idle`/`ready` vocabulary, whose `ready` names a storage-only child as resumable rather than terminal. Listing does not consult the continuation manager's Activation map, Agent registry, or provider availability; `send_message` remains the authoritative delivery-time operation, and a listed running continuable child may still reject delivery as an ownership conflict. The read-path rationale lives in [the list-identity-projection Agent Note](../../.agents/notes/implemented/architecture/2026-08-06-subagent-list-identity-projection.md).
 
 `SubagentService.listDescendants(rootSessionId)` applies the same live-preferred corpus and projection-backed interpretation to the root's complete descendant tree in stable pre-order. Ordinary sessions and one-shot children remain traversal nodes, so continuable descendants below them are discovered; only `origin: 'subagent'` candidates produce rows. Each returned child or diagnostic adds its position from the enumerated durable header, while a cold inspection revalidates that complete lifecycle before serving identity:
 
@@ -625,7 +646,7 @@ async start(name: string, request: SubagentStartRequest): Promise<SubagentRun>
 
 Types: [Agent](core.md) · [ContentBlock](llm-streaming.md) · [MessageId](llm-streaming.md) · [SessionId](core.md)
 
-Source: [`packages/subagent/subagent/src/index.ts:170`](../../packages/subagent/subagent/src/index.ts)
+Source: [`packages/subagent/subagent/src/index.ts:171`](../../packages/subagent/subagent/src/index.ts)
 
 <a id="subagent-events"></a>
 
@@ -651,7 +672,7 @@ A published child settled. Scope-filtered dispatch uses the same delegating pare
 
 Types: [Scoped](scope.md)
 
-Source: [`packages/subagent/subagent/src/index.ts:165`](../../packages/subagent/subagent/src/index.ts)
+Source: [`packages/subagent/subagent/src/index.ts:166`](../../packages/subagent/subagent/src/index.ts)
 
 <a id="subagentprovider-added--emit"></a>
 
@@ -668,7 +689,7 @@ A provider became resolvable in the registry.
 'subagent/provider-added'(provider: SubagentProvider): void
 ```
 
-Source: [`packages/subagent/subagent/src/index.ts:139`](../../packages/subagent/subagent/src/index.ts)
+Source: [`packages/subagent/subagent/src/index.ts:140`](../../packages/subagent/subagent/src/index.ts)
 
 <a id="subagentprovider-removed--emit"></a>
 
@@ -685,7 +706,7 @@ A provider left the registry. Accepted runs remain holder-owned.
 'subagent/provider-removed'(name: string): void
 ```
 
-Source: [`packages/subagent/subagent/src/index.ts:145`](../../packages/subagent/subagent/src/index.ts)
+Source: [`packages/subagent/subagent/src/index.ts:146`](../../packages/subagent/subagent/src/index.ts)
 
 <a id="subagentstart--emit"></a>
 
@@ -709,5 +730,5 @@ A provider established a published child. For in-process providers, `ctx.agents.
 
 Types: [Scoped](scope.md)
 
-Source: [`packages/subagent/subagent/src/index.ts:156`](../../packages/subagent/subagent/src/index.ts)
+Source: [`packages/subagent/subagent/src/index.ts:157`](../../packages/subagent/subagent/src/index.ts)
 <!-- END GENERATED cordis-surface -->
diff --git a/docs/subsystems/subagent.zh.md b/docs/subsystems/subagent.zh.md
index 1a8e2e5837..13754acc06 100644
--- a/docs/subsystems/subagent.zh.md
+++ b/docs/subsystems/subagent.zh.md
@@ -209,6 +209,27 @@ interface SubagentReportMessageSource {
 type SubagentReportDelivery = 'quiet' | 'wakeup'
 ```
 
+上报是 child 自己的选择，因此管理器还保有一份属于自己的记账：当驻留 Activation 结算时，它会向该 child 持久化的直接 parent 投递一条通知，说明该 epoch 如何结束，并携带其最终 assistant 内容。对每个调用方拿到过 id 的 child，这条投递都是无条件的；它发生在会让 parent 被判定为已结算的所有权释放之前，并通过与上报相同的唤醒准入记账到达驻留 parent。若 parent 自身所在的谱系已在拆卸中，这条通知会以不唤醒的方式送达，因为唤醒一个静息 Agent 是开启一个轮次，而不是排队等待工作。其来源信息使用一个独立的 kind，因此 transcript（文本记录）绝不会把运行时的记账呈现为 child 自己写下的内容。
+
+```ts type-equiv
+/**
+ * Durable attribution for the runtime's own account of a continuable child
+ * settling. Deliberately a different kind from
+ * {@link SubagentReportMessageSource}: a report is content the child chose,
+ * while this message is the manager stating what became of the child, and a
+ * transcript that merged them would credit the child with words it never wrote.
+ */
+interface SubagentSettledMessageSource {
+  readonly kind: 'subagent-settled'
+  /** A runtime account shown without expanding the row (`notice` context form). */
+  readonly form: 'notice'
+  /** One-line account of how the child ended. */
+  readonly summary: string
+  /** Session id of the child that settled. */
+  readonly senderSessionId: SessionId
+}
+```
+
 ```ts type-equiv
 /** Options for one continuable child's report to its direct parent. */
 interface SubagentReportOptions {
@@ -265,7 +286,7 @@ interface ContinuableCreateSpec {
 
 ## 持久化枚举：`listChildren()`、`listDescendants()` 与其条目
 
-`SubagentService.listChildren(parentSessionId)` 从 `ctx.sessions.list()` 与可选 `ctx.sessionPersistence.list()` 的实时优先合并中枚举 parent 直接且由会话支撑的 subagent——不经查询服务，也不会加载或恢复任何 Agent。候选是持久 header 携带 `origin: 'subagent'` 的直接 child；该标记只负责枚举分类与粗粒度的通用路由拒绝，不能证明描述符有效、child 可恢复或操作已获授权——身份由投影折叠负责，恢复由 Activation 约定负责。每行的 `mode`／`label` 是已注册 `subagent` projection unit 的值，经三级阶梯供值：存活 child 由注册表水位缓存供值（零日志读取）；冷 child 先读可选的投影 checkpoint 缓存（`cachedSnapshot`——过 own-suffix seq 门的身份即定值，own descriptor 一经追加不可变）；否则在一次 `persistence.inspect()` 读取上经注册表折叠（有界并发，每次列表重新计算）。该缓存是纯可选加速层：服务缺席、行里是 `null` 哨兵或 key 缺席、seq 门不过、读取出错，都静默落到权威重折。折叠规则是 `subagent/descriptor` last-wins 且没有失败通道：子 agent 自己的描述符覆盖 fork seed 中祖先的描述符，格式错误或版本不认识的载荷折叠为可序列化的 `null` 哨兵，视同无值。结果是按 `createdAt`、再按 id 排序的 `SubagentListEntry[]`：取到身份即生成带有 `mode: 'one-shot' | 'continuable'` 和 `activity: 'running' | 'inactive'` 的 `child` 条目；可继续条目始终携带 `label`，一次性条目则只在启动调用方提供展示元数据时携带该字段。已定局而折叠无身份的候选生成 `corrupt` diagnostic——缺失、格式错误与版本不认识的描述符有意不再细分（`unsupported` 仍保留在类型中但从不产出）；运行中而无身份的候选被省略（描述符落盘前的创建窗口）；冷检查失败生成一条 `unavailable` diagnostic 并在下次列表自然重试，因此一个损坏的 sibling 不会隐藏健康 child。`hasChildren` 标记存在持久 subagent origin 的直接后代，读取自同一份合并材料。活动状态只表示逻辑记录是否在 `ctx.sessions` 中存活，而不表示结果或可恢复性。缺少持久化时，枚举退化为仅存活枚举而不是报错——此时冷 child 本就无法恢复。缺少 `ctx.sessionProjections` 注册表时，`listChildren()` 抛出携带错误码 `SUBAGENT_CONTROL_PROJECTIONS_UNAVAILABLE` 的 `SubagentError`，缺少会话存储时则抛出 `SUBAGENT_CONTROL_SESSION_STORE_UNAVAILABLE`，两者都在任何读取之前检查，因此零 child 的部署同样确定失败；列表工具在插件加载时要求 `ctx.subagents` 与 `ctx.agents`。UI 等服务消费方可以展示两种模式，并为无标签的一次性 child 选择回退展示；面向模型的 `list_agents` 适配器（[dsh-tool-subagent-control](../../packages/subagent/tool-subagent-control) 中可单独加载的 `/list-agents` 插件）则只保留可继续条目，并通过在线 Agent 注册表将状态细化为 `running`／`idle`／`complete` 词汇。枚举不会查询继续执行管理器的 Activation map、Agent 注册表或提供方可用性；`send_message` 仍是消息送达时的权威操作，列表中的运行中可继续 child 仍可能因所有权冲突而拒绝投递。读路径的设计理由见[列表身份投影 Agent Note](../../.agents/notes/implemented/architecture/2026-08-06-subagent-list-identity-projection.md)。
+`SubagentService.listChildren(parentSessionId)` 从 `ctx.sessions.list()` 与可选 `ctx.sessionPersistence.list()` 的实时优先合并中枚举 parent 直接且由会话支撑的 subagent——不经查询服务，也不会加载或恢复任何 Agent。候选是持久 header 携带 `origin: 'subagent'` 的直接 child；该标记只负责枚举分类与粗粒度的通用路由拒绝，不能证明描述符有效、child 可恢复或操作已获授权——身份由投影折叠负责，恢复由 Activation 约定负责。每行的 `mode`／`label` 是已注册 `subagent` projection unit 的值，经三级阶梯供值：存活 child 由注册表水位缓存供值（零日志读取）；冷 child 先读可选的投影 checkpoint 缓存（`cachedSnapshot`——过 own-suffix seq 门的身份即定值，own descriptor 一经追加不可变）；否则在一次 `persistence.inspect()` 读取上经注册表折叠（有界并发，每次列表重新计算）。该缓存是纯可选加速层：服务缺席、行里是 `null` 哨兵或 key 缺席、seq 门不过、读取出错，都静默落到权威重折。折叠规则是 `subagent/descriptor` last-wins 且没有失败通道：子 agent 自己的描述符覆盖 fork seed 中祖先的描述符，格式错误或版本不认识的载荷折叠为可序列化的 `null` 哨兵，视同无值。结果是按 `createdAt`、再按 id 排序的 `SubagentListEntry[]`：取到身份即生成带有 `mode: 'one-shot' | 'continuable'` 和 `activity: 'running' | 'inactive'` 的 `child` 条目；可继续条目始终携带 `label`，一次性条目则只在启动调用方提供展示元数据时携带该字段。已定局而折叠无身份的候选生成 `corrupt` diagnostic——缺失、格式错误与版本不认识的描述符有意不再细分（`unsupported` 仍保留在类型中但从不产出）；运行中而无身份的候选被省略（描述符落盘前的创建窗口）；冷检查失败生成一条 `unavailable` diagnostic 并在下次列表自然重试，因此一个损坏的 sibling 不会隐藏健康 child。`hasChildren` 标记存在持久 subagent origin 的直接后代，读取自同一份合并材料。活动状态只表示逻辑记录是否在 `ctx.sessions` 中存活，而不表示结果或可恢复性。缺少持久化时，枚举退化为仅存活枚举而不是报错——此时冷 child 本就无法恢复。缺少 `ctx.sessionProjections` 注册表时，`listChildren()` 抛出携带错误码 `SUBAGENT_CONTROL_PROJECTIONS_UNAVAILABLE` 的 `SubagentError`，缺少会话存储时则抛出 `SUBAGENT_CONTROL_SESSION_STORE_UNAVAILABLE`，两者都在任何读取之前检查，因此零 child 的部署同样确定失败；列表工具在插件加载时要求 `ctx.subagents` 与 `ctx.agents`。UI 等服务消费方可以展示两种模式，并为无标签的一次性 child 选择回退展示；面向模型的 `list_agents` 适配器（[dsh-tool-subagent-control](../../packages/subagent/tool-subagent-control) 中可单独加载的 `/list-agents` 插件）则只保留可继续条目，并通过在线 Agent 注册表将状态细化为自己的 `running`／`idle`／`ready` 词汇，其中 `ready` 把仅存于存储的 child 命名为可恢复而非终态。枚举不会查询继续执行管理器的 Activation map、Agent 注册表或提供方可用性；`send_message` 仍是消息送达时的权威操作，列表中的运行中可继续 child 仍可能因所有权冲突而拒绝投递。读路径的设计理由见[列表身份投影 Agent Note](../../.agents/notes/implemented/architecture/2026-08-06-subagent-list-identity-projection.md)。
 
 `SubagentService.listDescendants(rootSessionId)` 将同一份实时优先语料与基于投影的解释应用到根的完整后代树，并按稳定 pre-order 输出。普通会话和一次性 child 仍作为遍历节点，因此其下的可继续后代仍可发现；只有 `origin: 'subagent'` 的候选会生成条目。每个返回的 child 或 diagnostic 都从枚举所得的持久 header 附加树位置；冷检查在提供身份前还会重新校验完整生命周期：
 
@@ -627,7 +648,7 @@ async start(name: string, request: SubagentStartRequest): Promise<SubagentRun>
 
 Types: [Agent](core.md) · [ContentBlock](llm-streaming.md) · [MessageId](llm-streaming.md) · [SessionId](core.md)
 
-Source: [`packages/subagent/subagent/src/index.ts:170`](../../packages/subagent/subagent/src/index.ts)
+Source: [`packages/subagent/subagent/src/index.ts:171`](../../packages/subagent/subagent/src/index.ts)
 
 <a id="subagent-events"></a>
 
@@ -653,7 +674,7 @@ A published child settled. Scope-filtered dispatch uses the same delegating pare
 
 Types: [Scoped](scope.md)
 
-Source: [`packages/subagent/subagent/src/index.ts:165`](../../packages/subagent/subagent/src/index.ts)
+Source: [`packages/subagent/subagent/src/index.ts:166`](../../packages/subagent/subagent/src/index.ts)
 
 <a id="subagentprovider-added--emit"></a>
 
@@ -670,7 +691,7 @@ A provider became resolvable in the registry.
 'subagent/provider-added'(provider: SubagentProvider): void
 ```
 
-Source: [`packages/subagent/subagent/src/index.ts:139`](../../packages/subagent/subagent/src/index.ts)
+Source: [`packages/subagent/subagent/src/index.ts:140`](../../packages/subagent/subagent/src/index.ts)
 
 <a id="subagentprovider-removed--emit"></a>
 
@@ -687,7 +708,7 @@ A provider left the registry. Accepted runs remain holder-owned.
 'subagent/provider-removed'(name: string): void
 ```
 
-Source: [`packages/subagent/subagent/src/index.ts:145`](../../packages/subagent/subagent/src/index.ts)
+Source: [`packages/subagent/subagent/src/index.ts:146`](../../packages/subagent/subagent/src/index.ts)
 
 <a id="subagentstart--emit"></a>
 
@@ -711,5 +732,5 @@ A provider established a published child. For in-process providers, `ctx.agents.
 
 Types: [Scoped](scope.md)
 
-Source: [`packages/subagent/subagent/src/index.ts:156`](../../packages/subagent/subagent/src/index.ts)
+Source: [`packages/subagent/subagent/src/index.ts:157`](../../packages/subagent/subagent/src/index.ts)
 <!-- END GENERATED cordis-surface -->
diff --git a/docs/tool-catalog.i18n.yaml b/docs/tool-catalog.i18n.yaml
index 7f4ebf012f..8dc9067615 100644
--- a/docs/tool-catalog.i18n.yaml
+++ b/docs/tool-catalog.i18n.yaml
@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write docs/tool-catalog.md
-tool-catalog.md: 19a4035fdc1e24d439307d11cb585f689b35043e
-tool-catalog.zh.md: e34fbb85152e012592b4e045e2f505ded6175e94
+tool-catalog.md: ca8335eecb8647e0741183f904c83ea03a72bb4e
+tool-catalog.zh.md: 4e446b0ae5568cea2c4ea2405467c24b158f3a4e
diff --git a/docs/tool-catalog.md b/docs/tool-catalog.md
index 19a4035fdc..ca8335eecb 100644
--- a/docs/tool-catalog.md
+++ b/docs/tool-catalog.md
@@ -31,9 +31,9 @@ This table connects model-visible tool names to the plugin package and service s
 | `@deepseek-ai/dsh-tool-ralph` | `ralph` | `ctx.tools`, `ctx.workflows`, `ctx.subagents`, `ctx.systemPrompt`, `a calling Agent (exec.agent parents every fresh round)` | `tool/call`, `tool/result`, `workflow and child session events during execution` | - | A fixed foreground workflow starts one fresh structured child per round; the model selects only the immutable objective and an optional round cap. |
 | `@deepseek-ai/dsh-tool-skill` | `skill` | `ctx.tools`, `ctx.agents`, `ctx.skills` | `tool/call`, `tool/result`, `user/message replacement catalogs via agent.inject()` | - | - |
 | `@deepseek-ai/dsh-tool-session-query` | `session_event_read`, `session_event_search`, `session_event_trace`, `session_search`, `session_trace` | `ctx.tools`, `ctx.systemPrompt`, `ctx.sessionQuery`, `a calling Agent for workspace authority` | `tool/call`, `tool/result` | - | The five read-only tools hide provider cursors and authorize every result from the immutable calling agent session. The package is opt-in; compositions that need enforced deadlines or bounded inline output also mount the generic timeout or spill policies. |
-| `@deepseek-ai/dsh-tool-subagent` | `subagent` | `ctx.tools`, `ctx.subagents` | `tool/call`, `tool/result`, `child session events through the chosen provider` | `subagent`, `subagent_fork` | The registered tool name is the load-time `toolName` config (default `subagent`); the schema above is that default. The shipped example agents load this package once per subagent backend, so the model additionally sees `subagent_fork` (bound to the fork backend) with an identical schema — see `packages/bundle/base/cordis.patch.yml` and `examples/acp-agent/cordis.yml`. |
+| `@deepseek-ai/dsh-tool-subagent` | `subagent` | `ctx.tools`, `ctx.subagents` | `tool/call`, `tool/result`, `child session events through the chosen provider` | `subagent`, `subagent_fork` | The registered tool name is the load-time `toolName` config (default `subagent`); the schema above is that default. The shipped compositions load this package once per subagent backend, so the model additionally sees `subagent_fork` bound to the fork backend. Each instance's description and `run_in_background` parameter follow its own `backgroundMode` and `enableRunInBackground`, so the two shipped schemas are not identical: `subagent` is `continuable`, while `subagent_fork` stays `one-shot` — see `packages/bundle/base/cordis.patch.yml` and `examples/acp-agent/cordis.yml`. |
 | `@deepseek-ai/dsh-tool-subagent-control` | `interrupt_agent`, `list_agents`, `send_message` | `ctx.tools`, `ctx.subagents`, `ctx.agents and ctx.sessionProjections (list_agents only)` | `tool/call`, `tool/result`, `child session events through ctx.subagents` | - | The globally named control tools over continuable background subagents: provider-bound `tool-subagent` instances register distinct delegation tools, while this package registers `send_message` and `interrupt_agent` once, plus `list_agents` from its separately loaded `/list-agents` plugin (whose catalog rows use the sessionProjections and live Agent registries). |
-| `@deepseek-ai/dsh-tool-subagent-report` | `report` | `ctx.subagents`, `a live continuable in-process child Agent` | `tool/call`, `tool/result`, `a user-role message in the direct parent session` | - | Registered per continuable in-process child rather than globally, so this schema is visible only inside such a child and survives its global `toolFilter`. The parent-facing `send_message` tool is installed independently. |
+| `@deepseek-ai/dsh-tool-subagent-report` | `report` | `ctx.subagents`, `ctx.systemPrompt`, `a live continuable in-process child Agent` | `tool/call`, `tool/result`, `a user-role message in the direct parent session` | - | Registered per continuable in-process child rather than globally, so this schema is visible only inside such a child and survives its global `toolFilter`. The same contribution installs the child-scoped `tool:report` prompt section, which this catalog does not render. The parent-facing `send_message` tool is installed independently. |
 | `@deepseek-ai/dsh-tool-tasks` | `task_kill`, `task_list`, `task_output` | `ctx.tools`, `ctx.tasks`, `ctx.systemPrompt` | `tool/call`, `tool/result`, `user/message via agent.inject() for background completion notices` | - | The kind-agnostic background-task control surface: background bash commands, PTY sends, and subagents are read, listed, and killed through the same three tools. Loading the plugin attaches the control surface that arms producers' `ctx.tasks.start()`. |
 | `@deepseek-ai/dsh-tool-todo` | `todo_write` | `ctx.tools`, `owning Agent session` | `tool/call`, `todo/write`, `tool/result` | - | todo_write is session-owned state; UIs render the latest todo/write event as a checklist. `allowParallelInProgress` is required with no default, so the catalog states its choice: `true`, whose description invites several `in_progress` items. A deployment choosing `false` receives the same tool with a description asking for exactly one active task. |
 | `@deepseek-ai/dsh-tool-workflow` | `workflow` | `ctx.tools`, `ctx.workflows`, `ctx.systemPrompt`, `a calling Agent (exec.agent parents the script children)` | `tool/call`, `tool/result` | - | - |
@@ -1210,7 +1210,7 @@ Delegate a self-contained task to a subagent (a separate agent that works in its
 
 Source: [`packages/subagent/tool-subagent/src/index.ts`](../packages/subagent/tool-subagent/src/index.ts)
 
-The registered tool name is the load-time `toolName` config (default `subagent`); the schema above is that default. The shipped example agents load this package once per subagent backend, so the model additionally sees `subagent_fork` (bound to the fork backend) with an identical schema — see `packages/bundle/base/cordis.patch.yml` and `examples/acp-agent/cordis.yml`.
+The registered tool name is the load-time `toolName` config (default `subagent`); the schema above is that default. The shipped compositions load this package once per subagent backend, so the model additionally sees `subagent_fork` bound to the fork backend. Each instance's description and `run_in_background` parameter follow its own `backgroundMode` and `enableRunInBackground`, so the two shipped schemas are not identical: `subagent` is `continuable`, while `subagent_fork` stays `one-shot` — see `packages/bundle/base/cordis.patch.yml` and `examples/acp-agent/cordis.yml`.
 
 ## `@deepseek-ai/dsh-tool-subagent-control`
 
@@ -1237,7 +1237,7 @@ Source: [`packages/subagent/tool-subagent-control/src/index.ts`](../packages/sub
 
 ### `list_agents`
 
-List your continuable background subagents by durable id and label. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and complete means it exists only in storage — a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.
+List your continuable background subagents by durable id and label. Use it to recall which ones you started, not to poll for completion — you are told when one finishes. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and ready means it exists only in storage — resumable, not terminal, and not a result waiting to be collected; a `send_message` starts a new turn on the same conversation, and a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.
 
 ```json
 {
@@ -1289,7 +1289,7 @@ The globally named control tools over continuable background subagents: provider
 
 ### `report`
 
-Report selected content to the agent that started you. Call this zero or more times for progress, findings, or a final answer. Reporting does not end your turn or finish your work, and only your direct parent receives it. A failed call may still have arrived, so do not blindly repeat it.
+Report selected content to the agent that started you. Call this once before you finish, with a self-contained final result, and earlier for progress or findings that change what that agent does next. That agent shares your workspace but does not automatically receive your transcript, tool output, or reasoning, so finishing your work is not itself a result. Reporting does not end your turn or finish your work, and only your direct parent receives it. A failed call may still have arrived, so do not blindly repeat it.
 
 ```json
 {
@@ -1297,7 +1297,7 @@ Report selected content to the agent that started you. Call this zero or more ti
   "properties": {
     "output": {
       "type": "string",
-      "description": "Self-contained content for your parent; it does not see your private work."
+      "description": "Actionable content for your parent; summarize conclusions and reference relevant shared paths."
     }
   },
   "required": [
@@ -1308,7 +1308,7 @@ Report selected content to the agent that started you. Call this zero or more ti
 
 Source: [`packages/subagent/tool-subagent-report/src/index.ts`](../packages/subagent/tool-subagent-report/src/index.ts)
 
-Registered per continuable in-process child rather than globally, so this schema is visible only inside such a child and survives its global `toolFilter`. The parent-facing `send_message` tool is installed independently.
+Registered per continuable in-process child rather than globally, so this schema is visible only inside such a child and survives its global `toolFilter`. The same contribution installs the child-scoped `tool:report` prompt section, which this catalog does not render. The parent-facing `send_message` tool is installed independently.
 
 ## `@deepseek-ai/dsh-tool-tasks`
 
diff --git a/docs/tool-catalog.zh.md b/docs/tool-catalog.zh.md
index e34fbb8515..4e446b0ae5 100644
--- a/docs/tool-catalog.zh.md
+++ b/docs/tool-catalog.zh.md
@@ -33,9 +33,9 @@
 | `@deepseek-ai/dsh-tool-ralph` | `ralph` | `ctx.tools`、`ctx.workflows`、`ctx.subagents`、`ctx.systemPrompt`、`a calling Agent (exec.agent parents every fresh round)` | `tool/call`、`tool/result`、`workflow and child session events during execution` | - | 固定的前台工作流会在每个 Round 启动一个全新的结构化子级；模型只能选择不可变目标和可选的 Round 上限。 |
 | `@deepseek-ai/dsh-tool-skill` | `skill` | `ctx.tools`、`ctx.agents`、`ctx.skills` | `tool/call`、`tool/result`、`user/message replacement catalogs via agent.inject()` | - | - |
 | `@deepseek-ai/dsh-tool-session-query` | `session_event_read`、`session_event_search`、`session_event_trace`、`session_search`、`session_trace` | `ctx.tools`、`ctx.systemPrompt`、`ctx.sessionQuery`、`a calling Agent for workspace authority` | `tool/call`、`tool/result` | - | 这 5 个只读工具会隐藏提供方游标，并根据不可变的调用 agent 会话为每个结果授权。该包需要选择启用；需要强制截止时间或限制行内输出的组合还会挂载通用超时或 spill 策略。 |
-| `@deepseek-ai/dsh-tool-subagent` | `subagent` | `ctx.tools`、`ctx.subagents` | `tool/call`、`tool/result`、`child session events through the chosen provider` | `subagent`、`subagent_fork` | 注册的工具名称取决于加载时 `toolName` 配置（默认为 `subagent`）；上述 schema 对应默认值。随产品发布的示例 agent 会为每个 subagent 后端加载一次该包，因此模型还会看到 schema 相同、绑定到 fork 后端的 `subagent_fork`；见 `packages/bundle/base/cordis.patch.yml` 和 `examples/acp-agent/cordis.yml`。 |
+| `@deepseek-ai/dsh-tool-subagent` | `subagent` | `ctx.tools`、`ctx.subagents` | `tool/call`、`tool/result`、`child session events through the chosen provider` | `subagent`、`subagent_fork` | 注册的工具名称取决于加载时 `toolName` 配置（默认为 `subagent`）；上述 schema 对应默认值。随产品发布的组合会为每个 subagent 后端加载一次该包，因此模型还会看到绑定到 fork 后端的 `subagent_fork`。每个实例的描述与 `run_in_background` 参数取决于它自己的 `backgroundMode` 与 `enableRunInBackground`，因此两个随附 schema 并不相同：`subagent` 为 `continuable`，而 `subagent_fork` 保持 `one-shot`；见 `packages/bundle/base/cordis.patch.yml` 和 `examples/acp-agent/cordis.yml`。 |
 | `@deepseek-ai/dsh-tool-subagent-control` | `interrupt_agent`、`list_agents`、`send_message` | `ctx.tools`、`ctx.subagents`、`ctx.agents and ctx.sessionProjections (list_agents only)` | `tool/call`、`tool/result`、`child session events through ctx.subagents` | - | 这些是控制可继续后台 subagent 的全局命名工具：绑定提供方的 `tool-subagent` 实例注册不同的委派工具；本包注册一次 `send_message` 和 `interrupt_agent`，另由 `list_agents` 通过单独加载的 `/list-agents` 插件提供，其目录行使用 sessionProjections 和实时 Agent 注册表。 |
-| `@deepseek-ai/dsh-tool-subagent-report` | `report` | `ctx.subagents`、`a live continuable in-process child Agent` | `tool/call`、`tool/result`、`a user-role message in the direct parent session` | - | 按可继续的进程内子级注册，而非全局注册，因此该 schema 仅在这种子级内部可见，并且不受其全局 `toolFilter` 影响。面向父级的 `send_message` 工具单独安装。 |
+| `@deepseek-ai/dsh-tool-subagent-report` | `report` | `ctx.subagents`、`ctx.systemPrompt`、`a live continuable in-process child Agent` | `tool/call`、`tool/result`、`a user-role message in the direct parent session` | - | 按可继续的进程内子级注册，而非全局注册，因此该 schema 仅在这种子级内部可见，并且不受其全局 `toolFilter` 影响。同一份贡献还会安装子级作用域的 `tool:report` 系统提示词 section，本目录不渲染该 section。面向父级的 `send_message` 工具单独安装。 |
 | `@deepseek-ai/dsh-tool-tasks` | `task_kill`、`task_list`、`task_output` | `ctx.tools`、`ctx.tasks`、`ctx.systemPrompt` | `tool/call`、`tool/result`、`user/message via agent.inject() for background completion notices` | - | 与任务种类无关的后台任务控制接口：后台 bash 命令、PTY 发送和 subagent 都通过相同的 3 个工具读取、列出和终止。加载该插件会挂接控制接口，从而启用生产方的 `ctx.tasks.start()`。 |
 | `@deepseek-ai/dsh-tool-todo` | `todo_write` | `ctx.tools`、`owning Agent session` | `tool/call`、`todo/write`、`tool/result` | - | todo_write 是会话所有的状态；UI 将最新的 todo/write 事件渲染为检查清单。`allowParallelInProgress` 是没有默认值的必填项，因此本目录明确选择 `true`，对应描述允许同时存在多个 `in_progress` 项。选择 `false` 的部署会获得同一工具，但描述会要求只能有 1 个活动任务。 |
 | `@deepseek-ai/dsh-tool-workflow` | `workflow` | `ctx.tools`、`ctx.workflows`、`ctx.systemPrompt`、`a calling Agent (exec.agent parents the script children)` | `tool/call`、`tool/result` | - | - |
@@ -1214,7 +1214,7 @@ lsp 工具将提供方选择和语言服务器子进程置于 ctx.lsp 之后，
 
 来源：[`packages/subagent/tool-subagent/src/index.ts`](../packages/subagent/tool-subagent/src/index.ts)
 
-注册的工具名称取决于加载时 `toolName` 配置（默认为 `subagent`）；上述 schema 对应默认值。随产品发布的示例 agent 会为每个 subagent 后端加载一次该包，因此模型还会看到 schema 相同、绑定到 fork 后端的 `subagent_fork`；见 `packages/bundle/base/cordis.patch.yml` 和 `examples/acp-agent/cordis.yml`。
+注册的工具名称取决于加载时 `toolName` 配置（默认为 `subagent`）；上述 schema 对应默认值。随产品发布的组合会为每个 subagent 后端加载一次该包，因此模型还会看到绑定到 fork 后端的 `subagent_fork`。每个实例的描述与 `run_in_background` 参数取决于它自己的 `backgroundMode` 与 `enableRunInBackground`，因此两个随附 schema 并不相同：`subagent` 为 `continuable`，而 `subagent_fork` 保持 `one-shot`；见 `packages/bundle/base/cordis.patch.yml` 和 `examples/acp-agent/cordis.yml`。
 
 ## `@deepseek-ai/dsh-tool-subagent-control`
 
@@ -1241,7 +1241,7 @@ lsp 工具将提供方选择和语言服务器子进程置于 ctx.lsp 之后，
 
 ### `list_agents`
 
-按持久 id 和标签列出你的可继续后台 subagent。状态来自实时注册表：running 表示 agent 此刻正在工作；idle 表示已加载但处于轮次之间，可能正在等待它启动的 agent；complete 表示它只存在于存储中。无论处于哪种状态，直接子级都仍可作为 `send_message` 的目标。该快照并非投递承诺；`send_message` 会执行权威检查，仍可能失败。无法读取的子级会作为诊断信息报告，而不会被静默丢弃。`descendants` 作用域会按稳定的前序顺序遍历你下方的整棵树，并为每个条目标注其持久的直接父会话 id 和深度。只有深度为 1 的条目可以使用 `send_message`；更深的条目只能作为 `interrupt_agent` 的候选目标。
+按持久 id 和标签列出你的可继续后台 subagent。用它回忆你启动过哪些 subagent，而不是轮询完成情况——subagent 完成时你会被告知。状态来自实时注册表：running 表示 agent 此刻正在工作；idle 表示已加载但处于轮次之间，可能正在等待它启动的 agent；ready 表示它只存在于存储中——可恢复而非终态，也不表示有结果等待收集；`send_message` 会在同一对话上开启新的轮次，且无论处于哪种状态，直接子级都仍可作为 `send_message` 的目标。该快照并非投递承诺；`send_message` 会执行权威检查，仍可能失败。无法读取的子级会作为诊断信息报告，而不会被静默丢弃。`descendants` 作用域会按稳定的前序顺序遍历你下方的整棵树，并为每个条目标注其持久的直接父会话 id 和深度。只有深度为 1 的条目可以使用 `send_message`；更深的条目只能作为 `interrupt_agent` 的候选目标。
 
 ```json
 {
@@ -1293,7 +1293,7 @@ lsp 工具将提供方选择和语言服务器子进程置于 ctx.lsp 之后，
 
 ### `report`
 
-向启动你的 agent 报告选定内容。你可以调用 0 次或多次，以报告进度、发现或最终答案。报告不会结束你的轮次或完成你的工作，且只有直接父级会收到。失败的调用仍可能已经送达，因此不要盲目重复。
+向启动你的 agent 报告选定内容。在你结束前调用一次，给出自包含的最终结果；当进度或发现会改变该 agent 接下来的行动时，也可以更早调用。该 agent 与你共享工作区，但不会自动收到你的 transcript（文本记录）、工具输出或推理，因此完成你的工作本身并不等于交出结果。报告不会结束你的轮次或完成你的工作，且只有直接父级会收到。失败的调用仍可能已经送达，因此不要盲目重复。
 
 ```json
 {
@@ -1301,7 +1301,7 @@ lsp 工具将提供方选择和语言服务器子进程置于 ctx.lsp 之后，
   "properties": {
     "output": {
       "type": "string",
-      "description": "Self-contained content for your parent; it does not see your private work."
+      "description": "Actionable content for your parent; summarize conclusions and reference relevant shared paths."
     }
   },
   "required": [
@@ -1312,7 +1312,7 @@ lsp 工具将提供方选择和语言服务器子进程置于 ctx.lsp 之后，
 
 来源：[`packages/subagent/tool-subagent-report/src/index.ts`](../packages/subagent/tool-subagent-report/src/index.ts)
 
-按可继续的进程内子级注册，而非全局注册，因此该 schema 仅在这种子级内部可见，并且不受其全局 `toolFilter` 影响。面向父级的 `send_message` 工具单独安装。
+按可继续的进程内子级注册，而非全局注册，因此该 schema 仅在这种子级内部可见，并且不受其全局 `toolFilter` 影响。同一份贡献还会安装子级作用域的 `tool:report` 系统提示词 section，本目录不渲染该 section。面向父级的 `send_message` 工具单独安装。
 
 ## `@deepseek-ai/dsh-tool-tasks`
 
diff --git a/examples/acp-agent/cordis.yml b/examples/acp-agent/cordis.yml
index 59e7b78278..a73d6d36e4 100644
--- a/examples/acp-agent/cordis.yml
+++ b/examples/acp-agent/cordis.yml
@@ -121,12 +121,16 @@
     backgroundMode: continuable
     maxDepth: 1
 
+# Fork stays one-shot because a continuable child's `report` tool and prompt
+# section precede the inherited history a fork reuses; `run_in_background` is off
+# because this example mounts no task service. See .agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md.
 - id: tool-subagent-fork
   name: '@deepseek-ai/dsh-tool-subagent'
   config:
     provider: fork
     toolName: subagent_fork
-    backgroundMode: continuable
+    backgroundMode: one-shot
+    enableRunInBackground: false
     maxDepth: 1
 
 
diff --git a/examples/acp-agent/subagent-report-quiet.cordis.snapshot.yml b/examples/acp-agent/subagent-report-quiet.cordis.snapshot.yml
new file mode 100644
index 0000000000..4d28b399f0
--- /dev/null
+++ b/examples/acp-agent/subagent-report-quiet.cordis.snapshot.yml
@@ -0,0 +1,51 @@
+# Keyless counterpart to subagent-report-quiet.cordis.yml: replace the live
+# adapter with replay, keep report delivery quiet, and fence the child behind
+# the end of its parent's spawn turn so settlement opens the next turn.
+- id: base
+  name: '@deepseek-ai/cordis-plugin-include'
+  config:
+    path: ./cordis.yml
+    patches:
+      - id: llm-deepseek
+        name: '@deepseek-ai/dsh-llm-deepseek'
+        disabled: true
+      - id: acp-agent
+        name: '@deepseek-ai/dsh-acp-demo'
+        config:
+          provider: deepseek-official
+          model: deepseek-v4-flash
+          persistenceRoot: !!js process.env.DSH_SNAPSHOT_SESSIONS_ROOT ?? './.sessions'
+          persistenceCompression: none
+          workspaceContext:
+            maxBytes: 65536
+          persona: |
+            You are a coding assistant powered by the {{model}} model. Your working directory is {{cwd}}. Your bash tool runs under a file sandbox — a `[sandbox: file access denied …]` result is policy, not a command bug.
+
+            Verify your work by running the code or tests. Keep answers brief and factual.
+      - id: sandbox
+        name: '@deepseek-ai/dsh-sandbox-local'
+        config:
+          runnerCommand:
+            - bash
+            - -c
+            - while [ "$1" != "--" ]; do shift; done; shift; exec "$@"
+            - passthrough-runner
+          runnerFailureSignatures:
+            - 'passthrough-runner: profile rejected'
+      - id: tool-subagent-report
+        name: '@deepseek-ai/dsh-tool-subagent-report'
+        config:
+          reportDelivery: quiet
+      - insert:
+          - id: llm-replay
+            name: '@deepseek-ai/dsh-llm-replay'
+            config:
+              providers:
+                - id: deepseek-official
+                  name: DeepSeek
+                  models:
+                    - id: deepseek-v4-flash
+                    - id: deepseek-v4-pro
+
+- id: report-fence
+  name: './tests/fixtures/subagent-report-fence.ts'
diff --git a/examples/acp-agent/subagent-report-quiet.cordis.yml b/examples/acp-agent/subagent-report-quiet.cordis.yml
new file mode 100644
index 0000000000..cb1c2b6d34
--- /dev/null
+++ b/examples/acp-agent/subagent-report-quiet.cordis.yml
@@ -0,0 +1,17 @@
+# Snapshot-only overlay pinning quiet report delivery. The shipped default wakes
+# the parent on every accepted report, and the runtime's settlement notice wakes
+# it again when the child's Activation ends; two independent wakes have no single
+# authored order. Quiet delivery leaves settlement as the only wake, while the
+# fixture below holds the child until the parent's spawn turn has closed.
+- id: base
+  name: '@deepseek-ai/cordis-plugin-include'
+  config:
+    path: ./cordis.yml
+    patches:
+      - id: tool-subagent-report
+        name: '@deepseek-ai/dsh-tool-subagent-report'
+        config:
+          reportDelivery: quiet
+
+- id: report-fence
+  name: './tests/fixtures/subagent-report-fence.ts'
diff --git a/examples/acp-agent/tests/acp.snapshot.ts b/examples/acp-agent/tests/acp.snapshot.ts
index 3ef3e71000..0cddc9a504 100644
--- a/examples/acp-agent/tests/acp.snapshot.ts
+++ b/examples/acp-agent/tests/acp.snapshot.ts
@@ -46,6 +46,9 @@ const CHILD_QUESTION_CONFIG = fileURLToPath(new URL('../child-question.cordis.ym
 const SESSION_SANDBOX_ROOT_CONFIG = fileURLToPath(new URL('../session-sandbox-root.cordis.yml', import.meta.url))
 const RETRY_CONFIG = fileURLToPath(new URL('../retry.cordis.yml', import.meta.url))
 const SESSION_TITLE_CONFIG = fileURLToPath(new URL('../session-title.cordis.yml', import.meta.url))
+const SUBAGENT_REPORT_QUIET_CONFIG = fileURLToPath(
+  new URL('../subagent-report-quiet.cordis.yml', import.meta.url),
+)
 const SUBAGENT_DURABILITY_FAILURE_CONFIG = fileURLToPath(
   new URL('../subagent-durability-failure.cordis.yml', import.meta.url),
 )
@@ -385,12 +388,19 @@ const SCENARIOS: Scenario[] = [
   // turns on that same child (the parent is never woken with their output),
   // send_message to an unknown subagent id fails without delivering, and the
   // child's retained handle is disposed child-first at teardown despite a
-  // failed final durability confirmation.
+  // failed final durability confirmation. That failed confirmation is also what
+  // the settlement notice must report: the child's last turn claimed the third
+  // message and then died on its durability checkpoint without entering a step,
+  // so the notice opening the parent's second turn says the child FAILED and the
+  // parent must not read the earlier answer as final. The scenario's fixture
+  // fences the child behind the parent's spawn turn so that notice can only
+  // arrive at an idle parent.
   {
     name: 'subagent-continuable',
     hasModelTurn: true,
     recorded: false,
     pinsChildToolSchemas: [1],
+    pinsChildSystemPrompts: [1],
     configPath: SUBAGENT_DURABILITY_FAILURE_CONFIG,
   },
   // Authored policy-inheritance transcript: the root session is switched to
@@ -398,11 +408,14 @@ const SCENARIOS: Scenario[] = [
   // continuable background child's log carries that override as a
   // `sandbox/mode` `source: 'delegation'` event, so the child's runtime
   // context states the inherited policy instead of the deployment default.
+  // The input also waits for the manager-owned settlement turn, keeping that
+  // delivery from racing transcript harvest.
   {
     name: 'subagent-continuable-inheritance',
     hasModelTurn: true,
     recorded: false,
     pinsChildToolSchemas: [1],
+    pinsChildSystemPrompts: [1],
     configPath: SUBAGENT_CONTINUABLE_INHERITANCE_CONFIG,
   },
   // The in-process child is published before its first follow-up fails. The
@@ -417,13 +430,18 @@ const SCENARIOS: Scenario[] = [
     configPath: SUBAGENT_DURABILITY_FAILURE_CONFIG,
   },
   // Authored child-to-parent transcript: the child calls its scope-local
-  // `report`, quiet delivery reaches the idle parent without waking it, and a
-  // later parent turn consumes the logged report.
+  // `report`, and the runtime's unconditional settlement notice then wakes the
+  // parked parent into one ordinary turn that claims both. The overlay pins
+  // quiet report delivery because two independent wakes have no orderable
+  // transcript; the shipped waking default is covered by package tests.
   {
     name: 'subagent-report',
     hasModelTurn: true,
     recorded: false,
+    overridden: false,
+    configPath: SUBAGENT_REPORT_QUIET_CONFIG,
     pinsChildToolSchemas: [1],
+    pinsChildSystemPrompts: [1],
   },
   // Authored durable-catalog transcript: the snapshot-only lifecycle marker
   // fences the second parent turn behind the child's Activation end, so
@@ -436,6 +454,7 @@ const SCENARIOS: Scenario[] = [
     hasModelTurn: true,
     recorded: false,
     pinsChildToolSchemas: [1],
+    pinsChildSystemPrompts: [1],
   },
   {
     name: 'subagent-depth-two-rejection',
diff --git a/examples/acp-agent/tests/fixtures/subagent-durability-failure.ts b/examples/acp-agent/tests/fixtures/subagent-durability-failure.ts
index 8f5185026e..9ebc0d6ac1 100644
--- a/examples/acp-agent/tests/fixtures/subagent-durability-failure.ts
+++ b/examples/acp-agent/tests/fixtures/subagent-durability-failure.ts
@@ -31,6 +31,8 @@ const FAILED_CHECKPOINT_TURN = 3
 /** Fail the child checkpoint and stabilize the authored follow-up failure ordering. */
 export function apply(ctx: Context): void {
   const followupsAccepted = Promise.withResolvers<undefined>()
+  const parentTurnClosed = Promise.withResolvers<undefined>()
+  let parentClosed = false
   const publishedFailure = process.env.DSH_SUBAGENT_PUBLISHED_FAILURE === '1'
   const persistence = ctx.sessionPersistence
   const load = persistence.load.bind(persistence)
@@ -62,8 +64,22 @@ export function apply(ctx: Context): void {
     agents.create = create
     persistence.load = load
     followupsAccepted.resolve(undefined)
+    parentTurnClosed.resolve(undefined)
   }, 'subagent snapshot ordering')
 
+  // The manager's settlement notice races whatever the parent is doing when the
+  // child's Activation ends, and this transcript pins it as the parent's own
+  // later turn. Hold the child's steps until the parent's spawn turn closes, so
+  // the notice can only arrive at an idle parent. The parent's turn never awaits
+  // child model work — its own fences need inbox acceptance only — so the child
+  // cannot deadlock it.
+  ctx.on('session/event', (session, event) => {
+    if (session.header.parentSession !== undefined || event.type !== 'turn/end') return
+    if (event.data.turn !== 1) return
+    parentClosed = true
+    parentTurnClosed.resolve(undefined)
+  })
+
   // Remap the placeholder child id in a follow-up to the live child. The child
   // id the model "knows" is authored into the transcript, while the running
   // child is minted with a random id, so without this the follow-ups would
@@ -91,7 +107,12 @@ export function apply(ctx: Context): void {
     if (accepted >= 3) followupsAccepted.resolve(undefined)
   })
   ctx.on('agent/pre-step', async ({ agent }, next) => {
-    if (agent.session.header.parentSession !== undefined) await followupsAccepted.promise
+    if (agent.session.header.parentSession === undefined) return next()
+    await followupsAccepted.promise
+    // The published-failure variant's child never reaches a step (its follow-up
+    // throws), and its parent turn awaits that child, so only the continuable
+    // scenario takes the settlement fence.
+    if (!publishedFailure && !parentClosed) await parentTurnClosed.promise
     return next()
   })
 
diff --git a/examples/acp-agent/tests/fixtures/subagent-report-fence.ts b/examples/acp-agent/tests/fixtures/subagent-report-fence.ts
new file mode 100644
index 0000000000..754343e4ec
--- /dev/null
+++ b/examples/acp-agent/tests/fixtures/subagent-report-fence.ts
@@ -0,0 +1,41 @@
+/**
+ * Loader fixture that holds the report child until its parent's spawn turn ends.
+ * @module subagent-report-fence
+ */
+
+import type { Context } from '@deepseek-ai/cordis'
+import type {} from '@deepseek-ai/dsh-agent-loop'
+
+/** Fixture plugin name. */
+export const name = 'subagent-report-fence'
+
+/**
+ * Keep replay scheduling from folding settlement into the parent's first turn.
+ * @param ctx - assembled ACP-agent context.
+ */
+export function apply(ctx: Context): void {
+  const childReady = Promise.withResolvers<undefined>()
+  const parentStopped = Promise.withResolvers<undefined>()
+  let hasStopped = false
+
+  ctx.effect(() => {
+    const disposeSession = ctx.root.on('session/event', (session, event) => {
+      if (session.header.parentSession !== undefined || event.type !== 'turn/end' || event.data.turn !== 1) return
+      hasStopped = true
+      parentStopped.resolve(undefined)
+    })
+    const disposeStep = ctx.root.on('agent/pre-step', async ({ agent, turn, step }, next) => {
+      if (agent.session.header.parentSession !== undefined) {
+        childReady.resolve(undefined)
+        if (!hasStopped) await parentStopped.promise
+      } else if (turn === 1 && step === 2) {
+        await childReady.promise
+      }
+      return next()
+    })
+    return () => {
+      disposeStep()
+      disposeSession()
+    }
+  }, 'subagent-report-fence.listeners')
+}
diff --git a/examples/acp-agent/tests/snapshots/advanced-toolchain/session.1.jsonl b/examples/acp-agent/tests/snapshots/advanced-toolchain/session.1.jsonl
index 4abda00a44..c5e1201fe8 100644
--- a/examples/acp-agent/tests/snapshots/advanced-toolchain/session.1.jsonl
+++ b/examples/acp-agent/tests/snapshots/advanced-toolchain/session.1.jsonl
@@ -6,7 +6,7 @@
 {"type":"subagent/descriptor","seq":4,"time":1786357538308,"data":{"version":2,"mode":"one-shot","provider":"spawn","label":"Check direct child"}}
 {"type":"step/start","seq":5,"time":1786357538310,"data":{"turn":1,"step":1}}
 {"type":"user/message","seq":6,"time":1785730458555,"data":{"content":[{"type":"text","text":"Reply with exactly DIRECT_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"ebe0cfa0-a909-47e0-8294-28ad84a8fe77"},"surfaceOp":"append"}
-{"type":"user/message","seq":7,"time":1786357538310,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"b8a6a626-fd2c-4602-a8d4-8074f229bfb5"},"surfaceOp":"append"}
+{"type":"user/message","seq":7,"time":1786357538310,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"3f5d0f63-fcf4-4b04-9c1f-6aabf60c8877"},"surfaceOp":"append"}
 {"type":"session/title","seq":8,"time":1786357538310,"data":{"title":"Reply with exactly DIRECT_CHILD_OK and","messageSeqs":[6],"source":{"kind":"fallback"}}}
 {"type":"request/header","seq":9,"time":1785730458555,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}}
 {"type":"request/context","seq":10,"time":1785730458555,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}}
diff --git a/examples/acp-agent/tests/snapshots/advanced-toolchain/session.2.jsonl b/examples/acp-agent/tests/snapshots/advanced-toolchain/session.2.jsonl
index 0a10247d2e..4d704dd8f2 100644
--- a/examples/acp-agent/tests/snapshots/advanced-toolchain/session.2.jsonl
+++ b/examples/acp-agent/tests/snapshots/advanced-toolchain/session.2.jsonl
@@ -6,7 +6,7 @@
 {"type":"subagent/descriptor","seq":4,"time":1786357538469,"data":{"version":2,"mode":"one-shot","provider":"spawn"}}
 {"type":"step/start","seq":5,"time":1786357538470,"data":{"turn":1,"step":1}}
 {"type":"user/message","seq":6,"time":1785730458703,"data":{"content":[{"type":"text","text":"Reply with exactly WORKFLOW_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"2ac2cc54-9bce-4cfa-a569-a64f51bc30a7"},"surfaceOp":"append"}
-{"type":"user/message","seq":7,"time":1786357538471,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"d327cdff-ad2e-4f9e-9c54-d4448ee11f2a"},"surfaceOp":"append"}
+{"type":"user/message","seq":7,"time":1786357538471,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"2bb5fd9d-ecb5-4597-ba8d-6626409278f5"},"surfaceOp":"append"}
 {"type":"session/title","seq":8,"time":1786357538471,"data":{"title":"Reply with exactly WORKFLOW_CHILD_OK and","messageSeqs":[6],"source":{"kind":"fallback"}}}
 {"type":"request/header","seq":9,"time":1785730458703,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}}
 {"type":"request/context","seq":10,"time":1785730458703,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}}
diff --git a/examples/acp-agent/tests/snapshots/advanced-toolchain/session.jsonl b/examples/acp-agent/tests/snapshots/advanced-toolchain/session.jsonl
index aea9e3107c..56ae9d71c6 100644
--- a/examples/acp-agent/tests/snapshots/advanced-toolchain/session.jsonl
+++ b/examples/acp-agent/tests/snapshots/advanced-toolchain/session.jsonl
@@ -4,7 +4,7 @@
 {"type":"agent/inbox/spliced","seq":2,"time":1785821417919,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}
 {"type":"step/start","seq":3,"time":1783957884486,"data":{"turn":1,"step":1}}
 {"type":"user/message","seq":4,"time":1785498801761,"data":{"content":[{"type":"text","text":"Run this advanced flow exactly once: try a no-op temporary Cordis Plugin named snapshot-marker; use run_code to inspect the live temporary Plugins through tools.cordis_inspect; delegate once to a direct spawn child; run one workflow that delegates to another spawn child; stop dyn-1; then reply with exactly ADVANCED_ACP_OK."}],"source":{"kind":"user"},"role":"user","id":"6e45782a-31be-4ba7-8c4a-7411a2027e36"},"surfaceOp":"append"}
-{"type":"user/message","seq":5,"time":1785730458430,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}]},"role":"user","id":"06416873-c855-452d-8996-ea5cf45223d1"},"surfaceOp":"append"}
+{"type":"user/message","seq":5,"time":1785730458430,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}]},"role":"user","id":"b6de532e-08ff-42b0-abae-1895dc463500"},"surfaceOp":"append"}
 {"type":"session/title","seq":6,"time":1785730458430,"data":{"title":"Run this advanced flow exactly","messageSeqs":[4],"source":{"kind":"fallback"}}}
 {"type":"request/header","seq":7,"time":1785498801765,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}}
 {"type":"request/context","seq":8,"time":1785730458431,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}}
diff --git a/examples/acp-agent/tests/snapshots/advanced-toolchain/system-prompt.expected.md b/examples/acp-agent/tests/snapshots/advanced-toolchain/system-prompt.expected.md
index b613fbac76..91ecf6c888 100644
--- a/examples/acp-agent/tests/snapshots/advanced-toolchain/system-prompt.expected.md
+++ b/examples/acp-agent/tests/snapshots/advanced-toolchain/system-prompt.expected.md
@@ -99,7 +99,7 @@ interface ToolArgsMap {
     /** The agent id of the running agent to interrupt. */
     agent_id: string;
   } & Record<string, JsonValue>;
-  /** List your continuable background subagents by durable id and label. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and complete means it exists only in storage — a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only. */
+  /** List your continuable background subagents by durable id and label. Use it to recall which ones you started, not to poll for completion — you are told when one finishes. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and ready means it exists only in storage — resumable, not terminal, and not a result waiting to be collected; a `send_message` starts a new turn on the same conversation, and a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only. */
   list_agents: {
     /** children (default) lists direct children only; descendants walks the complete tree below you. */
     scope?: "children" | "descendants";
@@ -132,23 +132,21 @@ interface ToolArgsMap {
     /** The exact skill name from the available skills list. */
     name: string;
   } & Record<string, JsonValue>;
-  /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */
+  /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work. */
   subagent: {
     /** A short (3-5 word) description of the delegated task, for display. */
     description: string;
     /** The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs. */
     prompt: string;
-    /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */
+    /** Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message. */
     run_in_background?: boolean;
   } & Record<string, JsonValue>;
-  /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */
+  /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. */
   subagent_fork: {
     /** A short (3-5 word) description of the delegated task, for display. */
     description: string;
     /** The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new. */
     prompt: string;
-    /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */
-    run_in_background?: boolean;
   } & Record<string, JsonValue>;
   /** Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops. */
   task_kill: {
@@ -319,7 +317,7 @@ interface ToolOutputMap {
     kind: "child";
     id: string;
     label: string;
-    status: "running" | "idle" | "complete";
+    status: "running" | "idle" | "ready";
     parent?: string;
     depth?: number;
   } | {
diff --git a/examples/acp-agent/tests/snapshots/advanced-toolchain/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/advanced-toolchain/tool-schemas.expected.json
index c979e1db87..339fac4f73 100644
--- a/examples/acp-agent/tests/snapshots/advanced-toolchain/tool-schemas.expected.json
+++ b/examples/acp-agent/tests/snapshots/advanced-toolchain/tool-schemas.expected.json
@@ -190,7 +190,7 @@
     },
     {
       "name": "list_agents",
-      "description": "List your continuable background subagents by durable id and label. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and complete means it exists only in storage — a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
+      "description": "List your continuable background subagents by durable id and label. Use it to recall which ones you started, not to poll for completion — you are told when one finishes. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and ready means it exists only in storage — resumable, not terminal, and not a result waiting to be collected; a `send_message` starts a new turn on the same conversation, and a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -309,7 +309,7 @@
     },
     {
       "name": "subagent",
-      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -323,7 +323,7 @@
           },
           "run_in_background": {
             "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
+            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."
           }
         },
         "required": [
@@ -334,7 +334,7 @@
     },
     {
       "name": "subagent_fork",
-      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -345,10 +345,6 @@
           "prompt": {
             "type": "string",
             "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."
-          },
-          "run_in_background": {
-            "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
           }
         },
         "required": [
diff --git a/examples/acp-agent/tests/snapshots/both-mode-turn/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/both-mode-turn/tool-schemas.expected.json
index d5cbdc2ccc..597408965a 100644
--- a/examples/acp-agent/tests/snapshots/both-mode-turn/tool-schemas.expected.json
+++ b/examples/acp-agent/tests/snapshots/both-mode-turn/tool-schemas.expected.json
@@ -133,7 +133,7 @@
     },
     {
       "name": "list_agents",
-      "description": "List your continuable background subagents by durable id and label. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and complete means it exists only in storage — a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
+      "description": "List your continuable background subagents by durable id and label. Use it to recall which ones you started, not to poll for completion — you are told when one finishes. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and ready means it exists only in storage — resumable, not terminal, and not a result waiting to be collected; a `send_message` starts a new turn on the same conversation, and a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -252,7 +252,7 @@
     },
     {
       "name": "subagent",
-      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -266,7 +266,7 @@
           },
           "run_in_background": {
             "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
+            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."
           }
         },
         "required": [
@@ -277,7 +277,7 @@
     },
     {
       "name": "subagent_fork",
-      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -288,10 +288,6 @@
           "prompt": {
             "type": "string",
             "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."
-          },
-          "run_in_background": {
-            "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
           }
         },
         "required": [
diff --git a/examples/acp-agent/tests/snapshots/code-mode-turn/system-prompt.expected.md b/examples/acp-agent/tests/snapshots/code-mode-turn/system-prompt.expected.md
index cb1b047e93..94e7ee4afb 100644
--- a/examples/acp-agent/tests/snapshots/code-mode-turn/system-prompt.expected.md
+++ b/examples/acp-agent/tests/snapshots/code-mode-turn/system-prompt.expected.md
@@ -82,7 +82,7 @@ interface ToolArgsMap {
     /** The agent id of the running agent to interrupt. */
     agent_id: string;
   } & Record<string, JsonValue>;
-  /** List your continuable background subagents by durable id and label. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and complete means it exists only in storage — a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only. */
+  /** List your continuable background subagents by durable id and label. Use it to recall which ones you started, not to poll for completion — you are told when one finishes. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and ready means it exists only in storage — resumable, not terminal, and not a result waiting to be collected; a `send_message` starts a new turn on the same conversation, and a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only. */
   list_agents: {
     /** children (default) lists direct children only; descendants walks the complete tree below you. */
     scope?: "children" | "descendants";
@@ -115,23 +115,21 @@ interface ToolArgsMap {
     /** The exact skill name from the available skills list. */
     name: string;
   } & Record<string, JsonValue>;
-  /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */
+  /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work. */
   subagent: {
     /** A short (3-5 word) description of the delegated task, for display. */
     description: string;
     /** The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs. */
     prompt: string;
-    /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */
+    /** Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message. */
     run_in_background?: boolean;
   } & Record<string, JsonValue>;
-  /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */
+  /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. */
   subagent_fork: {
     /** A short (3-5 word) description of the delegated task, for display. */
     description: string;
     /** The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new. */
     prompt: string;
-    /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */
-    run_in_background?: boolean;
   } & Record<string, JsonValue>;
   /** Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops. */
   task_kill: {
@@ -290,7 +288,7 @@ interface ToolOutputMap {
     kind: "child";
     id: string;
     label: string;
-    status: "running" | "idle" | "complete";
+    status: "running" | "idle" | "ready";
     parent?: string;
     depth?: number;
   } | {
diff --git a/examples/acp-agent/tests/snapshots/lsp-definition/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/lsp-definition/tool-schemas.expected.json
index 335be3672b..d1cf4a4ef6 100644
--- a/examples/acp-agent/tests/snapshots/lsp-definition/tool-schemas.expected.json
+++ b/examples/acp-agent/tests/snapshots/lsp-definition/tool-schemas.expected.json
@@ -133,7 +133,7 @@
     },
     {
       "name": "list_agents",
-      "description": "List your continuable background subagents by durable id and label. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and complete means it exists only in storage — a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
+      "description": "List your continuable background subagents by durable id and label. Use it to recall which ones you started, not to poll for completion — you are told when one finishes. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and ready means it exists only in storage — resumable, not terminal, and not a result waiting to be collected; a `send_message` starts a new turn on the same conversation, and a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -268,7 +268,7 @@
     },
     {
       "name": "subagent",
-      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -282,7 +282,7 @@
           },
           "run_in_background": {
             "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
+            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."
           }
         },
         "required": [
@@ -293,7 +293,7 @@
     },
     {
       "name": "subagent_fork",
-      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -304,10 +304,6 @@
           "prompt": {
             "type": "string",
             "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."
-          },
-          "run_in_background": {
-            "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
           }
         },
         "required": [
diff --git a/examples/acp-agent/tests/snapshots/product-subagent-both/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/product-subagent-both/tool-schemas.expected.json
index 76f60e28d4..6274c67e7f 100644
--- a/examples/acp-agent/tests/snapshots/product-subagent-both/tool-schemas.expected.json
+++ b/examples/acp-agent/tests/snapshots/product-subagent-both/tool-schemas.expected.json
@@ -133,7 +133,7 @@
     },
     {
       "name": "list_agents",
-      "description": "List your continuable background subagents by durable id and label. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and complete means it exists only in storage — a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
+      "description": "List your continuable background subagents by durable id and label. Use it to recall which ones you started, not to poll for completion — you are told when one finishes. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and ready means it exists only in storage — resumable, not terminal, and not a result waiting to be collected; a `send_message` starts a new turn on the same conversation, and a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -231,7 +231,7 @@
     },
     {
       "name": "subagent",
-      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -245,7 +245,7 @@
           },
           "run_in_background": {
             "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
+            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."
           }
         },
         "required": [
@@ -298,7 +298,7 @@
     },
     {
       "name": "subagent_fork",
-      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -309,10 +309,6 @@
           "prompt": {
             "type": "string",
             "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."
-          },
-          "run_in_background": {
-            "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
           }
         },
         "required": [
diff --git a/examples/acp-agent/tests/snapshots/product-subagent-codex/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/product-subagent-codex/tool-schemas.expected.json
index 84c4671579..e7e83d948c 100644
--- a/examples/acp-agent/tests/snapshots/product-subagent-codex/tool-schemas.expected.json
+++ b/examples/acp-agent/tests/snapshots/product-subagent-codex/tool-schemas.expected.json
@@ -133,7 +133,7 @@
     },
     {
       "name": "list_agents",
-      "description": "List your continuable background subagents by durable id and label. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and complete means it exists only in storage — a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
+      "description": "List your continuable background subagents by durable id and label. Use it to recall which ones you started, not to poll for completion — you are told when one finishes. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and ready means it exists only in storage — resumable, not terminal, and not a result waiting to be collected; a `send_message` starts a new turn on the same conversation, and a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -231,7 +231,7 @@
     },
     {
       "name": "subagent",
-      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -245,7 +245,7 @@
           },
           "run_in_background": {
             "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
+            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."
           }
         },
         "required": [
@@ -277,7 +277,7 @@
     },
     {
       "name": "subagent_fork",
-      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -288,10 +288,6 @@
           "prompt": {
             "type": "string",
             "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."
-          },
-          "run_in_background": {
-            "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
           }
         },
         "required": [
diff --git a/examples/acp-agent/tests/snapshots/pty-tools/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/pty-tools/tool-schemas.expected.json
index 37d7d467be..272fe439d3 100644
--- a/examples/acp-agent/tests/snapshots/pty-tools/tool-schemas.expected.json
+++ b/examples/acp-agent/tests/snapshots/pty-tools/tool-schemas.expected.json
@@ -133,7 +133,7 @@
     },
     {
       "name": "list_agents",
-      "description": "List your continuable background subagents by durable id and label. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and complete means it exists only in storage — a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
+      "description": "List your continuable background subagents by durable id and label. Use it to recall which ones you started, not to poll for completion — you are told when one finishes. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and ready means it exists only in storage — resumable, not terminal, and not a result waiting to be collected; a `send_message` starts a new turn on the same conversation, and a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -231,7 +231,7 @@
     },
     {
       "name": "subagent",
-      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -245,7 +245,7 @@
           },
           "run_in_background": {
             "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
+            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."
           }
         },
         "required": [
@@ -256,7 +256,7 @@
     },
     {
       "name": "subagent_fork",
-      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -267,10 +267,6 @@
           "prompt": {
             "type": "string",
             "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."
-          },
-          "run_in_background": {
-            "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
           }
         },
         "required": [
diff --git a/examples/acp-agent/tests/snapshots/read-image/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/read-image/tool-schemas.expected.json
index 8da6396169..0b495864d6 100644
--- a/examples/acp-agent/tests/snapshots/read-image/tool-schemas.expected.json
+++ b/examples/acp-agent/tests/snapshots/read-image/tool-schemas.expected.json
@@ -133,7 +133,7 @@
     },
     {
       "name": "list_agents",
-      "description": "List your continuable background subagents by durable id and label. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and complete means it exists only in storage — a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
+      "description": "List your continuable background subagents by durable id and label. Use it to recall which ones you started, not to poll for completion — you are told when one finishes. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and ready means it exists only in storage — resumable, not terminal, and not a result waiting to be collected; a `send_message` starts a new turn on the same conversation, and a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -247,7 +247,7 @@
     },
     {
       "name": "subagent",
-      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -261,7 +261,7 @@
           },
           "run_in_background": {
             "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
+            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."
           }
         },
         "required": [
@@ -272,7 +272,7 @@
     },
     {
       "name": "subagent_fork",
-      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -283,10 +283,6 @@
           "prompt": {
             "type": "string",
             "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."
-          },
-          "run_in_background": {
-            "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
           }
         },
         "required": [
diff --git a/examples/acp-agent/tests/snapshots/session-query-spill/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/session-query-spill/tool-schemas.expected.json
index aed34842df..f37db3d88b 100644
--- a/examples/acp-agent/tests/snapshots/session-query-spill/tool-schemas.expected.json
+++ b/examples/acp-agent/tests/snapshots/session-query-spill/tool-schemas.expected.json
@@ -133,7 +133,7 @@
     },
     {
       "name": "list_agents",
-      "description": "List your continuable background subagents by durable id and label. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and complete means it exists only in storage — a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
+      "description": "List your continuable background subagents by durable id and label. Use it to recall which ones you started, not to poll for completion — you are told when one finishes. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and ready means it exists only in storage — resumable, not terminal, and not a result waiting to be collected; a `send_message` starts a new turn on the same conversation, and a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -435,7 +435,7 @@
     },
     {
       "name": "subagent",
-      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -449,7 +449,7 @@
           },
           "run_in_background": {
             "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
+            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."
           }
         },
         "required": [
@@ -460,7 +460,7 @@
     },
     {
       "name": "subagent_fork",
-      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -471,10 +471,6 @@
           "prompt": {
             "type": "string",
             "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."
-          },
-          "run_in_background": {
-            "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
           }
         },
         "required": [
diff --git a/examples/acp-agent/tests/snapshots/subagent-child-question-rejection/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/subagent-child-question-rejection/tool-schemas.expected.json
index dc07c990aa..ffd8b4bd7e 100644
--- a/examples/acp-agent/tests/snapshots/subagent-child-question-rejection/tool-schemas.expected.json
+++ b/examples/acp-agent/tests/snapshots/subagent-child-question-rejection/tool-schemas.expected.json
@@ -196,7 +196,7 @@
     },
     {
       "name": "list_agents",
-      "description": "List your continuable background subagents by durable id and label. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and complete means it exists only in storage — a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
+      "description": "List your continuable background subagents by durable id and label. Use it to recall which ones you started, not to poll for completion — you are told when one finishes. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and ready means it exists only in storage — resumable, not terminal, and not a result waiting to be collected; a `send_message` starts a new turn on the same conversation, and a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -294,7 +294,7 @@
     },
     {
       "name": "subagent",
-      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -308,7 +308,7 @@
           },
           "run_in_background": {
             "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
+            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."
           }
         },
         "required": [
@@ -319,7 +319,7 @@
     },
     {
       "name": "subagent_fork",
-      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -330,10 +330,6 @@
           "prompt": {
             "type": "string",
             "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."
-          },
-          "run_in_background": {
-            "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
           }
         },
         "required": [
diff --git a/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/input.json b/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/input.json
index 183e21b557..f07f1d51fc 100644
--- a/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/input.json
+++ b/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/input.json
@@ -14,6 +14,13 @@
    "op": "waitForSubagentTurnEnd",
    "child": 1,
    "minimumTurn": 1
+  },
+  {
+   "op": "waitForTurnStart",
+   "minimumTurn": 2
+  },
+  {
+   "op": "waitForTurnEnd"
   }
  ]
 }
diff --git a/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/session.jsonl b/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/session.jsonl
index 15b1748ea5..293f9c3b83 100644
--- a/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/session.jsonl
+++ b/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/session.jsonl
@@ -27,3 +27,16 @@
 {"type":"assistant/message","seq":25,"time":1786333735904,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"text","text":"DONE"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"4057a08e-b50e-45e7-beb0-c74485f2b7d6"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[20,21,22,23,24],"surfaceOp":"append"}
 {"type":"step/end","seq":26,"time":1786333735904,"data":{"turn":1,"step":2}}
 {"type":"turn/end","seq":27,"time":1786333735904,"data":{"turn":1,"reason":{"kind":"completed"}}}
+{"type":"agent/inbox/spliced","seq":28,"time":1786374357751,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more."},{"type":"text","text":"Its closing message:"},{"type":"text","text":"CHILD_OK"}],"source":{"kind":"subagent-settled","form":"notice","summary":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more.","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"e0bd4902-daba-4e23-bfcb-9e102fdd203d"}]}}
+{"type":"turn/start","seq":29,"time":1786374357751,"data":{"turn":2}}
+{"type":"agent/inbox/spliced","seq":30,"time":1786374357751,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}
+{"type":"step/start","seq":31,"time":1786374357758,"data":{"turn":2,"step":1}}
+{"type":"user/message","seq":32,"time":1786374357758,"data":{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more."},{"type":"text","text":"Its closing message:"},{"type":"text","text":"CHILD_OK"}],"source":{"kind":"subagent-settled","form":"notice","summary":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more.","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"e0bd4902-daba-4e23-bfcb-9e102fdd203d"},"surfaceOp":"append"}
+{"type":"assistant/chunk","seq":33,"time":1786374357764,"data":{"turn":2,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
+{"type":"assistant/chunk","seq":34,"time":1786374357764,"data":{"turn":2,"step":1,"chunk":{"type":"text-delta","index":0,"text":"SUBAGENT_SETTLED_NOTED"}}}
+{"type":"assistant/chunk","seq":35,"time":1786374357764,"data":{"turn":2,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}}}}
+{"type":"assistant/chunk","seq":36,"time":1786374357764,"data":{"turn":2,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
+{"type":"assistant/chunk","seq":37,"time":1786374357764,"data":{"turn":2,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}
+{"type":"assistant/message","seq":38,"time":1786374357764,"data":{"turn":2,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"bbe5ef7b-2a3a-47f4-8475-60945b31a373"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[33,34,35,36,37],"surfaceOp":"append"}
+{"type":"step/end","seq":39,"time":1786374357765,"data":{"turn":2,"step":1}}
+{"type":"turn/end","seq":40,"time":1786374357765,"data":{"turn":2,"reason":{"kind":"completed"}}}
diff --git a/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/stdout.expected.jsonl b/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/stdout.expected.jsonl
index 82ae8907ca..d6a2728b5a 100644
--- a/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/stdout.expected.jsonl
+++ b/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/stdout.expected.jsonl
@@ -2,3 +2,4 @@
 {"jsonrpc":"2.0","id":2,"result":{"sessionId":"{{sessionId}}"}}
 {"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"DONE"}}}}
 {"jsonrpc":"2.0","id":3,"result":{"stopReason":"end_turn"}}
+{"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}}}}
diff --git a/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/system-prompt.1.expected.md b/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/system-prompt.1.expected.md
new file mode 100644
index 0000000000..71f4ed05f8
--- /dev/null
+++ b/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/system-prompt.1.expected.md
@@ -0,0 +1,24 @@
+You are an AI agent powered by the DeepSeek Harness SDK.
+
+You are a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}. Your bash tool runs under a file sandbox — a `[sandbox: file access denied …]` result is policy, not a command bug.
+
+Verify your work by running the code or tests. Keep answers brief and factual.
+
+
+Use the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.
+
+Use the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.
+
+Use the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.
+
+Check the [exit code: N] marker on every bash result; investigate failures before moving on.
+
+Track every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.
+
+Use goal tools for one long-running completion objective in the current session. create_goal may infer goal intent from a direct human request in any language; do not create a goal for routine single-turn work. Call get_goal before update_goal and copy its exact goal_id and revision. After session resume or fork, an active goal is disarmed: when a human asks to continue or resume in any wording or language, use update_goal action resume to rearm it. Mark complete only when the objective is actually achieved. Mark blocked only after the same blocking condition persists for at least 3 consecutive goal rounds, and report that concrete condition in blocked_reason; difficulty, uncertainty, or useful remaining work is not blocked.
+
+Use the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.
+
+Use the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.
+
+Deliver your result with the report tool before you finish: call it once with a self-contained answer. The agent that started you shares your workspace but does not automatically receive your transcript, tool output, or reasoning, so a closing remark such as "done" leaves it nothing it can use. Report earlier as well whenever a partial finding changes what that agent should do next; reporting never ends your turn.
diff --git a/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/tool-schemas.1.expected.json b/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/tool-schemas.1.expected.json
index 7dd791cf27..253fbc4a32 100644
--- a/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/tool-schemas.1.expected.json
+++ b/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/tool-schemas.1.expected.json
@@ -133,7 +133,7 @@
     },
     {
       "name": "list_agents",
-      "description": "List your continuable background subagents by durable id and label. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and complete means it exists only in storage — a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
+      "description": "List your continuable background subagents by durable id and label. Use it to recall which ones you started, not to poll for completion — you are told when one finishes. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and ready means it exists only in storage — resumable, not terminal, and not a result waiting to be collected; a `send_message` starts a new turn on the same conversation, and a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -194,13 +194,13 @@
     },
     {
       "name": "report",
-      "description": "Report selected content to the agent that started you. Call this zero or more times for progress, findings, or a final answer. Reporting does not end your turn or finish your work, and only your direct parent receives it. A failed call may still have arrived, so do not blindly repeat it.",
+      "description": "Report selected content to the agent that started you. Call this once before you finish, with a self-contained final result, and earlier for progress or findings that change what that agent does next. That agent shares your workspace but does not automatically receive your transcript, tool output, or reasoning, so finishing your work is not itself a result. Reporting does not end your turn or finish your work, and only your direct parent receives it. A failed call may still have arrived, so do not blindly repeat it.",
       "parameters": {
         "type": "object",
         "properties": {
           "output": {
             "type": "string",
-            "description": "Self-contained content for your parent; it does not see your private work."
+            "description": "Actionable content for your parent; summarize conclusions and reference relevant shared paths."
           }
         },
         "required": [
@@ -247,7 +247,7 @@
     },
     {
       "name": "subagent",
-      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -261,7 +261,7 @@
           },
           "run_in_background": {
             "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
+            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."
           }
         },
         "required": [
@@ -272,7 +272,7 @@
     },
     {
       "name": "subagent_fork",
-      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -283,10 +283,6 @@
           "prompt": {
             "type": "string",
             "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."
-          },
-          "run_in_background": {
-            "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
           }
         },
         "required": [
diff --git a/examples/acp-agent/tests/snapshots/subagent-continuable/input.json b/examples/acp-agent/tests/snapshots/subagent-continuable/input.json
index f1354bd76a..380a7825c2 100644
--- a/examples/acp-agent/tests/snapshots/subagent-continuable/input.json
+++ b/examples/acp-agent/tests/snapshots/subagent-continuable/input.json
@@ -10,6 +10,17 @@
       "op": "prompt",
       "text": "Follow these steps exactly, then stop. 1. Call the subagent tool once with run_in_background set to true, description 'Reply with CHILD_OK', and prompt 'Reply with exactly the word CHILD_OK and nothing else.'. 2. Call send_message twice in a row, both with the subagent id from step 1: first with message 'Now reply with exactly SECOND_OK.', then with message 'Now reply with exactly THIRD_OK.'. 3. Call send_message with subagent_id exactly '22222222-2222-4222-8222-222222222222' (a subagent that does not exist) and message 'Please continue.', and observe that it fails. 4. Reply with the single word DONE. Do not use the bash tool."
     },
-    { "op": "waitForSubagentTurnEnd", "child": 1, "minimumTurn": 3 }
+    {
+      "op": "waitForSubagentTurnEnd",
+      "child": 1,
+      "minimumTurn": 3
+    },
+    {
+      "op": "waitForTurnStart",
+      "minimumTurn": 2
+    },
+    {
+      "op": "waitForTurnEnd"
+    }
   ]
 }
diff --git a/examples/acp-agent/tests/snapshots/subagent-continuable/session.jsonl b/examples/acp-agent/tests/snapshots/subagent-continuable/session.jsonl
index a759c35a32..60a55d1373 100644
--- a/examples/acp-agent/tests/snapshots/subagent-continuable/session.jsonl
+++ b/examples/acp-agent/tests/snapshots/subagent-continuable/session.jsonl
@@ -1,10 +1,10 @@
 {"type":"session","version":0,"id":"11111111-1111-4111-8111-111111111111","createdAt":1789000000000,"cwd":"{{cwd}}","delegationDepth":0}
-{"type":"agent/inbox/spliced","seq":0,"time":1785730451297,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Follow these steps exactly, then stop. 1. Call the subagent tool once with run_in_background set to true, description 'Reply with CHILD_OK', and prompt 'Reply with exactly the word CHILD_OK and nothing else.'. 2. Call send_message twice in a row, both with the subagent id from step 1: first with message 'Now reply with exactly SECOND_OK.', then with message 'Now reply with exactly THIRD_OK.'. 3. Call send_message with subagent_id exactly '22222222-2222-4222-8222-222222222222' (a subagent that does not exist) and message 'Please continue.', and observe that it fails. 4. Reply with the single word DONE. Do not use the bash tool."}],"source":{"kind":"user"},"role":"user","id":"125665d3-8c03-4190-b4f9-c27d61d245f4"}]}}
+{"type":"agent/inbox/spliced","seq":0,"time":1785730451297,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Follow these steps exactly, then stop. 1. Call the subagent tool once with run_in_background set to true, description 'Reply with CHILD_OK', and prompt 'Reply with exactly the word CHILD_OK and nothing else.'. 2. Call send_message twice in a row, both with the subagent id from step 1: first with message 'Now reply with exactly SECOND_OK.', then with message 'Now reply with exactly THIRD_OK.'. 3. Call send_message with subagent_id exactly '22222222-2222-4222-8222-222222222222' (a subagent that does not exist) and message 'Please continue.', and observe that it fails. 4. Reply with the single word DONE. Do not use the bash tool."}],"source":{"kind":"user"},"role":"user","id":"579d3d6d-a57e-4d55-9b48-05832a79d9f8"}]}}
 {"type":"turn/start","seq":1,"time":1785821408972,"data":{"turn":1}}
 {"type":"agent/inbox/spliced","seq":2,"time":1785821408972,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}
 {"type":"step/start","seq":3,"time":1785730451327,"data":{"turn":1,"step":1}}
-{"type":"user/message","seq":4,"time":1785730451327,"data":{"content":[{"type":"text","text":"Follow these steps exactly, then stop. 1. Call the subagent tool once with run_in_background set to true, description 'Reply with CHILD_OK', and prompt 'Reply with exactly the word CHILD_OK and nothing else.'. 2. Call send_message twice in a row, both with the subagent id from step 1: first with message 'Now reply with exactly SECOND_OK.', then with message 'Now reply with exactly THIRD_OK.'. 3. Call send_message with subagent_id exactly '22222222-2222-4222-8222-222222222222' (a subagent that does not exist) and message 'Please continue.', and observe that it fails. 4. Reply with the single word DONE. Do not use the bash tool."}],"source":{"kind":"user"},"role":"user","id":"125665d3-8c03-4190-b4f9-c27d61d245f4"},"surfaceOp":"append"}
-{"type":"user/message","seq":5,"time":1785730451328,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}]},"role":"user","id":"533e9513-a329-4a36-9a8d-ddaf544b57c3"},"surfaceOp":"append"}
+{"type":"user/message","seq":4,"time":1785730451327,"data":{"content":[{"type":"text","text":"Follow these steps exactly, then stop. 1. Call the subagent tool once with run_in_background set to true, description 'Reply with CHILD_OK', and prompt 'Reply with exactly the word CHILD_OK and nothing else.'. 2. Call send_message twice in a row, both with the subagent id from step 1: first with message 'Now reply with exactly SECOND_OK.', then with message 'Now reply with exactly THIRD_OK.'. 3. Call send_message with subagent_id exactly '22222222-2222-4222-8222-222222222222' (a subagent that does not exist) and message 'Please continue.', and observe that it fails. 4. Reply with the single word DONE. Do not use the bash tool."}],"source":{"kind":"user"},"role":"user","id":"579d3d6d-a57e-4d55-9b48-05832a79d9f8"},"surfaceOp":"append"}
+{"type":"user/message","seq":5,"time":1785730451328,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}]},"role":"user","id":"c4f5f7ed-1c11-4f31-923f-3142c79f0c2c"},"surfaceOp":"append"}
 {"type":"session/title","seq":6,"time":1785730451328,"data":{"title":"Follow these steps exactly, then","messageSeqs":[4],"source":{"kind":"fallback"}}}
 {"type":"request/header","seq":7,"time":1785730451329,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}}
 {"type":"request/context","seq":8,"time":1785730451329,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}}
@@ -13,9 +13,9 @@
 {"type":"assistant/chunk","seq":11,"time":1785544945188,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Reply with CHILD_OK\", \"prompt\": \"Reply with exactly the word CHILD_OK and nothing else.\", \"run_in_background\": true}"}}}}
 {"type":"assistant/chunk","seq":12,"time":1785544945188,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
 {"type":"assistant/chunk","seq":13,"time":1785730451338,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
-{"type":"assistant/message","seq":14,"time":1785730451338,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Reply with CHILD_OK\", \"prompt\": \"Reply with exactly the word CHILD_OK and nothing else.\", \"run_in_background\": true}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"4a2c2f8b-66be-4860-9bbf-b83feb56009e"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[9,10,11,12,13],"surfaceOp":"append"}
+{"type":"assistant/message","seq":14,"time":1785730451338,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Reply with CHILD_OK\", \"prompt\": \"Reply with exactly the word CHILD_OK and nothing else.\", \"run_in_background\": true}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"680da987-6d29-4141-b83d-af57b050c712"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[9,10,11,12,13],"surfaceOp":"append"}
 {"type":"tool/call","seq":15,"time":1785730451338,"data":{"turn":1,"step":1,"callId":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Reply with CHILD_OK\", \"prompt\": \"Reply with exactly the word CHILD_OK and nothing else.\", \"run_in_background\": true}"}}
-{"type":"tool/result","seq":16,"time":1785730451348,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_bg_start"},"content":[{"type":"tool-result","toolCallId":"call_bg_start","content":[{"type":"text","text":"started subagent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"b54a0d61-4233-40f1-ab3a-3eb58e1b0c61"}},"sourceEventSeqs":[15],"surfaceOp":"append"}
+{"type":"tool/result","seq":16,"time":1785730451348,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_bg_start"},"content":[{"type":"tool-result","toolCallId":"call_bg_start","content":[{"type":"text","text":"started subagent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"7825edb2-080e-49c1-ba74-ad69d16bf566"}},"sourceEventSeqs":[15],"surfaceOp":"append"}
 {"type":"step/end","seq":17,"time":1785730451348,"data":{"turn":1,"step":1}}
 {"type":"step/start","seq":18,"time":1785730451360,"data":{"turn":1,"step":2}}
 {"type":"assistant/chunk","seq":19,"time":1789000000019,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}
@@ -23,9 +23,9 @@
 {"type":"assistant/chunk","seq":21,"time":1785544945212,"data":{"turn":1,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"call_followup_1","name":"send_message","arguments":"{\"subagent_id\": \"33333333-3333-4333-8333-333333333333\", \"message\": \"Now reply with exactly SECOND_OK.\"}"}}}}
 {"type":"assistant/chunk","seq":22,"time":1785544945212,"data":{"turn":1,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
 {"type":"assistant/chunk","seq":23,"time":1785730451364,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
-{"type":"assistant/message","seq":24,"time":1785730451364,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_followup_1","name":"send_message","arguments":"{\"subagent_id\": \"33333333-3333-4333-8333-333333333333\", \"message\": \"Now reply with exactly SECOND_OK.\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"398dea92-a100-4b2f-a9e1-72629def132d"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"}
+{"type":"assistant/message","seq":24,"time":1785730451364,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_followup_1","name":"send_message","arguments":"{\"subagent_id\": \"33333333-3333-4333-8333-333333333333\", \"message\": \"Now reply with exactly SECOND_OK.\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"ef6eadc7-165e-4705-b865-3889f0af0f36"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"}
 {"type":"tool/call","seq":25,"time":1785730451365,"data":{"turn":1,"step":2,"callId":"call_followup_1","name":"send_message","arguments":"{\"subagent_id\": \"33333333-3333-4333-8333-333333333333\", \"message\": \"Now reply with exactly SECOND_OK.\"}"}}
-{"type":"tool/result","seq":26,"time":1785730451377,"data":{"turn":1,"step":2,"message":{"source":{"kind":"tool","callId":"call_followup_1"},"content":[{"type":"tool-result","toolCallId":"call_followup_1","content":[{"type":"text","text":"message queued as the next turn for subagent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"45b388fd-6a48-4a02-9f3b-1d642e797c71"}},"sourceEventSeqs":[25],"surfaceOp":"append"}
+{"type":"tool/result","seq":26,"time":1785730451377,"data":{"turn":1,"step":2,"message":{"source":{"kind":"tool","callId":"call_followup_1"},"content":[{"type":"tool-result","toolCallId":"call_followup_1","content":[{"type":"text","text":"message queued as the next turn for subagent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"ac1214a5-1d91-4fab-8f96-833baca114f8"}},"sourceEventSeqs":[25],"surfaceOp":"append"}
 {"type":"step/end","seq":27,"time":1785730451377,"data":{"turn":1,"step":2}}
 {"type":"step/start","seq":28,"time":1785730451390,"data":{"turn":1,"step":3}}
 {"type":"assistant/chunk","seq":29,"time":1789000000029,"data":{"turn":1,"step":3,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}
@@ -33,9 +33,9 @@
 {"type":"assistant/chunk","seq":31,"time":1785544945241,"data":{"turn":1,"step":3,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"call_followup_2","name":"send_message","arguments":"{\"subagent_id\": \"33333333-3333-4333-8333-333333333333\", \"message\": \"Now reply with exactly THIRD_OK.\"}"}}}}
 {"type":"assistant/chunk","seq":32,"time":1785544945242,"data":{"turn":1,"step":3,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
 {"type":"assistant/chunk","seq":33,"time":1785730451394,"data":{"turn":1,"step":3,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
-{"type":"assistant/message","seq":34,"time":1785730451394,"data":{"turn":1,"step":3,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_followup_2","name":"send_message","arguments":"{\"subagent_id\": \"33333333-3333-4333-8333-333333333333\", \"message\": \"Now reply with exactly THIRD_OK.\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"f56cec19-e761-4c77-9237-07d12d334275"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[29,30,31,32,33],"surfaceOp":"append"}
+{"type":"assistant/message","seq":34,"time":1785730451394,"data":{"turn":1,"step":3,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_followup_2","name":"send_message","arguments":"{\"subagent_id\": \"33333333-3333-4333-8333-333333333333\", \"message\": \"Now reply with exactly THIRD_OK.\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"33939813-0792-4ac5-8864-ec62a4ddff8e"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[29,30,31,32,33],"surfaceOp":"append"}
 {"type":"tool/call","seq":35,"time":1785730451395,"data":{"turn":1,"step":3,"callId":"call_followup_2","name":"send_message","arguments":"{\"subagent_id\": \"33333333-3333-4333-8333-333333333333\", \"message\": \"Now reply with exactly THIRD_OK.\"}"}}
-{"type":"tool/result","seq":36,"time":1785730451406,"data":{"turn":1,"step":3,"message":{"source":{"kind":"tool","callId":"call_followup_2"},"content":[{"type":"tool-result","toolCallId":"call_followup_2","content":[{"type":"text","text":"message queued as the next turn for subagent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"193df31b-de9a-4723-b2b1-c7c96a8eaee4"}},"sourceEventSeqs":[35],"surfaceOp":"append"}
+{"type":"tool/result","seq":36,"time":1785730451406,"data":{"turn":1,"step":3,"message":{"source":{"kind":"tool","callId":"call_followup_2"},"content":[{"type":"tool-result","toolCallId":"call_followup_2","content":[{"type":"text","text":"message queued as the next turn for subagent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"a6f64c64-f50c-47cd-a6b3-a3b57d3dc83d"}},"sourceEventSeqs":[35],"surfaceOp":"append"}
 {"type":"step/end","seq":37,"time":1785730451406,"data":{"turn":1,"step":3}}
 {"type":"step/start","seq":38,"time":1785730451419,"data":{"turn":1,"step":4}}
 {"type":"assistant/chunk","seq":39,"time":1789000000039,"data":{"turn":1,"step":4,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}
@@ -43,9 +43,9 @@
 {"type":"assistant/chunk","seq":41,"time":1785544945273,"data":{"turn":1,"step":4,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"call_followup_unknown","name":"send_message","arguments":"{\"subagent_id\": \"22222222-2222-4222-8222-222222222222\", \"message\": \"Please continue.\"}"}}}}
 {"type":"assistant/chunk","seq":42,"time":1785544945273,"data":{"turn":1,"step":4,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
 {"type":"assistant/chunk","seq":43,"time":1785730451424,"data":{"turn":1,"step":4,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
-{"type":"assistant/message","seq":44,"time":1785730451424,"data":{"turn":1,"step":4,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_followup_unknown","name":"send_message","arguments":"{\"subagent_id\": \"22222222-2222-4222-8222-222222222222\", \"message\": \"Please continue.\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"cbf54e23-fb96-45cc-b629-b9cb48fc9876"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[39,40,41,42,43],"surfaceOp":"append"}
+{"type":"assistant/message","seq":44,"time":1785730451424,"data":{"turn":1,"step":4,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_followup_unknown","name":"send_message","arguments":"{\"subagent_id\": \"22222222-2222-4222-8222-222222222222\", \"message\": \"Please continue.\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"95eab91d-b103-4033-8e2e-c9c93b1b0211"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[39,40,41,42,43],"surfaceOp":"append"}
 {"type":"tool/call","seq":45,"time":1785730451425,"data":{"turn":1,"step":4,"callId":"call_followup_unknown","name":"send_message","arguments":"{\"subagent_id\": \"22222222-2222-4222-8222-222222222222\", \"message\": \"Please continue.\"}"}}
-{"type":"tool/result","seq":46,"time":1785730451437,"data":{"turn":1,"step":4,"message":{"source":{"kind":"tool","callId":"call_followup_unknown"},"content":[{"type":"tool-result","toolCallId":"call_followup_unknown","content":[{"type":"text","text":"Error: subagent \"22222222-2222-4222-8222-222222222222\" is unavailable"}],"isError":true}],"role":"user","id":"d5d961af-c171-40b6-87a1-33d2c14740a7"},"error":{"name":"SubagentError","code":"NOT_RESUMABLE"}},"sourceEventSeqs":[45],"surfaceOp":"append"}
+{"type":"tool/result","seq":46,"time":1785730451437,"data":{"turn":1,"step":4,"message":{"source":{"kind":"tool","callId":"call_followup_unknown"},"content":[{"type":"tool-result","toolCallId":"call_followup_unknown","content":[{"type":"text","text":"Error: subagent \"22222222-2222-4222-8222-222222222222\" is unavailable"}],"isError":true}],"role":"user","id":"8a095e4b-3059-420d-856f-1cbd20b6a2e2"},"error":{"name":"SubagentError","code":"NOT_RESUMABLE"}},"sourceEventSeqs":[45],"surfaceOp":"append"}
 {"type":"step/end","seq":47,"time":1785730451437,"data":{"turn":1,"step":4}}
 {"type":"step/start","seq":48,"time":1785730451450,"data":{"turn":1,"step":5}}
 {"type":"assistant/chunk","seq":49,"time":1785394678779,"data":{"turn":1,"step":5,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
@@ -53,6 +53,19 @@
 {"type":"assistant/chunk","seq":51,"time":1785544945303,"data":{"turn":1,"step":5,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"DONE"}}}}
 {"type":"assistant/chunk","seq":52,"time":1785544945303,"data":{"turn":1,"step":5,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
 {"type":"assistant/chunk","seq":53,"time":1785730451453,"data":{"turn":1,"step":5,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}
-{"type":"assistant/message","seq":54,"time":1785730451453,"data":{"turn":1,"step":5,"message":{"role":"assistant","content":[{"type":"text","text":"DONE"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"fbea6b64-84cf-4411-abec-48d26b3801da"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[49,50,51,52,53],"surfaceOp":"append"}
+{"type":"assistant/message","seq":54,"time":1785730451453,"data":{"turn":1,"step":5,"message":{"role":"assistant","content":[{"type":"text","text":"DONE"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"eb51ecb3-3347-4216-ad4e-c2130c43ecfc"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[49,50,51,52,53],"surfaceOp":"append"}
 {"type":"step/end","seq":55,"time":1785730451454,"data":{"turn":1,"step":5}}
 {"type":"turn/end","seq":56,"time":1785730451454,"data":{"turn":1,"reason":{"kind":"completed"}}}
+{"type":"agent/inbox/spliced","seq":57,"time":1786011346660,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 failed before it finished."},{"type":"text","text":"Its closing message:"},{"type":"text","text":"SECOND_OK"}],"source":{"kind":"subagent-settled","form":"notice","summary":"Background subagent 33333333-3333-4333-8333-333333333333 failed before it finished.","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"2cf0afd2-ee6e-4a3f-a35a-2fd4d5b665ca"}]}}
+{"type":"turn/start","seq":58,"time":1786011346660,"data":{"turn":2}}
+{"type":"agent/inbox/spliced","seq":59,"time":1786011346660,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}
+{"type":"step/start","seq":60,"time":1786012466689,"data":{"turn":2,"step":1}}
+{"type":"user/message","seq":61,"time":1786012466689,"data":{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 failed before it finished."},{"type":"text","text":"Its closing message:"},{"type":"text","text":"SECOND_OK"}],"source":{"kind":"subagent-settled","form":"notice","summary":"Background subagent 33333333-3333-4333-8333-333333333333 failed before it finished.","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"2cf0afd2-ee6e-4a3f-a35a-2fd4d5b665ca"},"surfaceOp":"append"}
+{"type":"assistant/chunk","seq":62,"time":1786012466693,"data":{"turn":2,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
+{"type":"assistant/chunk","seq":63,"time":1786012466693,"data":{"turn":2,"step":1,"chunk":{"type":"text-delta","index":0,"text":"SUBAGENT_SETTLED_NOTED"}}}
+{"type":"assistant/chunk","seq":64,"time":1786012466693,"data":{"turn":2,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}}}}
+{"type":"assistant/chunk","seq":65,"time":1786012466693,"data":{"turn":2,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
+{"type":"assistant/chunk","seq":66,"time":1786012466693,"data":{"turn":2,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}
+{"type":"assistant/message","seq":67,"time":1786333283620,"data":{"turn":2,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"758adcee-9284-4889-86a2-0181a278a754"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[62,63,64,65,66],"surfaceOp":"append"}
+{"type":"step/end","seq":68,"time":1786333283620,"data":{"turn":2,"step":1}}
+{"type":"turn/end","seq":69,"time":1786333283620,"data":{"turn":2,"reason":{"kind":"completed"}}}
diff --git a/examples/acp-agent/tests/snapshots/subagent-continuable/stdout.expected.jsonl b/examples/acp-agent/tests/snapshots/subagent-continuable/stdout.expected.jsonl
index 82ae8907ca..d6a2728b5a 100644
--- a/examples/acp-agent/tests/snapshots/subagent-continuable/stdout.expected.jsonl
+++ b/examples/acp-agent/tests/snapshots/subagent-continuable/stdout.expected.jsonl
@@ -2,3 +2,4 @@
 {"jsonrpc":"2.0","id":2,"result":{"sessionId":"{{sessionId}}"}}
 {"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"DONE"}}}}
 {"jsonrpc":"2.0","id":3,"result":{"stopReason":"end_turn"}}
+{"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}}}}
diff --git a/examples/acp-agent/tests/snapshots/subagent-continuable/system-prompt.1.expected.md b/examples/acp-agent/tests/snapshots/subagent-continuable/system-prompt.1.expected.md
new file mode 100644
index 0000000000..71f4ed05f8
--- /dev/null
+++ b/examples/acp-agent/tests/snapshots/subagent-continuable/system-prompt.1.expected.md
@@ -0,0 +1,24 @@
+You are an AI agent powered by the DeepSeek Harness SDK.
+
+You are a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}. Your bash tool runs under a file sandbox — a `[sandbox: file access denied …]` result is policy, not a command bug.
+
+Verify your work by running the code or tests. Keep answers brief and factual.
+
+
+Use the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.
+
+Use the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.
+
+Use the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.
+
+Check the [exit code: N] marker on every bash result; investigate failures before moving on.
+
+Track every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.
+
+Use goal tools for one long-running completion objective in the current session. create_goal may infer goal intent from a direct human request in any language; do not create a goal for routine single-turn work. Call get_goal before update_goal and copy its exact goal_id and revision. After session resume or fork, an active goal is disarmed: when a human asks to continue or resume in any wording or language, use update_goal action resume to rearm it. Mark complete only when the objective is actually achieved. Mark blocked only after the same blocking condition persists for at least 3 consecutive goal rounds, and report that concrete condition in blocked_reason; difficulty, uncertainty, or useful remaining work is not blocked.
+
+Use the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.
+
+Use the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.
+
+Deliver your result with the report tool before you finish: call it once with a self-contained answer. The agent that started you shares your workspace but does not automatically receive your transcript, tool output, or reasoning, so a closing remark such as "done" leaves it nothing it can use. Report earlier as well whenever a partial finding changes what that agent should do next; reporting never ends your turn.
diff --git a/examples/acp-agent/tests/snapshots/subagent-continuable/tool-schemas.1.expected.json b/examples/acp-agent/tests/snapshots/subagent-continuable/tool-schemas.1.expected.json
index 7dd791cf27..253fbc4a32 100644
--- a/examples/acp-agent/tests/snapshots/subagent-continuable/tool-schemas.1.expected.json
+++ b/examples/acp-agent/tests/snapshots/subagent-continuable/tool-schemas.1.expected.json
@@ -133,7 +133,7 @@
     },
     {
       "name": "list_agents",
-      "description": "List your continuable background subagents by durable id and label. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and complete means it exists only in storage — a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
+      "description": "List your continuable background subagents by durable id and label. Use it to recall which ones you started, not to poll for completion — you are told when one finishes. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and ready means it exists only in storage — resumable, not terminal, and not a result waiting to be collected; a `send_message` starts a new turn on the same conversation, and a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -194,13 +194,13 @@
     },
     {
       "name": "report",
-      "description": "Report selected content to the agent that started you. Call this zero or more times for progress, findings, or a final answer. Reporting does not end your turn or finish your work, and only your direct parent receives it. A failed call may still have arrived, so do not blindly repeat it.",
+      "description": "Report selected content to the agent that started you. Call this once before you finish, with a self-contained final result, and earlier for progress or findings that change what that agent does next. That agent shares your workspace but does not automatically receive your transcript, tool output, or reasoning, so finishing your work is not itself a result. Reporting does not end your turn or finish your work, and only your direct parent receives it. A failed call may still have arrived, so do not blindly repeat it.",
       "parameters": {
         "type": "object",
         "properties": {
           "output": {
             "type": "string",
-            "description": "Self-contained content for your parent; it does not see your private work."
+            "description": "Actionable content for your parent; summarize conclusions and reference relevant shared paths."
           }
         },
         "required": [
@@ -247,7 +247,7 @@
     },
     {
       "name": "subagent",
-      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -261,7 +261,7 @@
           },
           "run_in_background": {
             "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
+            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."
           }
         },
         "required": [
@@ -272,7 +272,7 @@
     },
     {
       "name": "subagent_fork",
-      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -283,10 +283,6 @@
           "prompt": {
             "type": "string",
             "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."
-          },
-          "run_in_background": {
-            "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
           }
         },
         "required": [
diff --git a/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.1.jsonl b/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.1.jsonl
index 44d587c851..d793d62624 100644
--- a/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.1.jsonl
+++ b/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.1.jsonl
@@ -6,7 +6,7 @@
 {"type":"subagent/descriptor","seq":4,"time":1786357533600,"data":{"version":2,"mode":"one-shot","provider":"spawn","label":"Start depth one"}}
 {"type":"step/start","seq":5,"time":1786357533602,"data":{"turn":1,"step":1}}
 {"type":"user/message","seq":6,"time":1785730456014,"data":{"content":[{"type":"text","text":"Call subagent once. Ask that child to attempt one further subagent call, then report the result."}],"source":{"kind":"user"},"role":"user","id":"a8129357-1bde-4cbd-90b4-6b8ad51d52e1"},"surfaceOp":"append"}
-{"type":"user/message","seq":7,"time":1786357533602,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"82a11d62-da49-4ad3-a243-789ea3cd7c08"},"surfaceOp":"append"}
+{"type":"user/message","seq":7,"time":1786357533602,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"d8d6b5bf-db7b-484f-8095-e35915b94274"},"surfaceOp":"append"}
 {"type":"session/title","seq":8,"time":1786357533602,"data":{"title":"Call subagent once. Ask that","messageSeqs":[6],"source":{"kind":"fallback"}}}
 {"type":"request/header","seq":9,"time":1785730456014,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}}
 {"type":"request/context","seq":10,"time":1785730456014,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}}
diff --git a/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.2.jsonl b/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.2.jsonl
index 89de9b1c4f..62c259289d 100644
--- a/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.2.jsonl
+++ b/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.2.jsonl
@@ -6,7 +6,7 @@
 {"type":"subagent/descriptor","seq":4,"time":1786357533628,"data":{"version":2,"mode":"one-shot","provider":"spawn","label":"Start depth two"}}
 {"type":"step/start","seq":5,"time":1786357533630,"data":{"turn":1,"step":1}}
 {"type":"user/message","seq":6,"time":1785730456041,"data":{"content":[{"type":"text","text":"Attempt one subagent call beyond the configured cap, then report the rejection."}],"source":{"kind":"user"},"role":"user","id":"d4dc5a16-e542-4dd9-8e82-e6b7829cfc4b"},"surfaceOp":"append"}
-{"type":"user/message","seq":7,"time":1786357533630,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"d5488efe-eea2-4019-8fcb-7e6e49077d8a"},"surfaceOp":"append"}
+{"type":"user/message","seq":7,"time":1786357533630,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"e9907a11-3d7c-4796-a459-16afd4afde32"},"surfaceOp":"append"}
 {"type":"session/title","seq":8,"time":1786357533630,"data":{"title":"Attempt one subagent call beyond","messageSeqs":[6],"source":{"kind":"fallback"}}}
 {"type":"request/header","seq":9,"time":1785730456041,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}}
 {"type":"request/context","seq":10,"time":1785730456042,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}}
diff --git a/examples/acp-agent/tests/snapshots/subagent-list-agents/input.json b/examples/acp-agent/tests/snapshots/subagent-list-agents/input.json
index b3b1d8e170..ab0c00f2f2 100644
--- a/examples/acp-agent/tests/snapshots/subagent-list-agents/input.json
+++ b/examples/acp-agent/tests/snapshots/subagent-list-agents/input.json
@@ -14,6 +14,13 @@
       "op": "waitForFile",
       "path": ".dsh-snapshot-subagent-settled"
     },
+    {
+      "op": "waitForTurnStart",
+      "minimumTurn": 2
+    },
+    {
+      "op": "waitForTurnEnd"
+    },
     {
       "op": "prompt",
       "text": "Call list_agents once with scope set to descendants and observe the subagent you started. Then call interrupt_agent once with agent_id set to 33333333-3333-4333-8333-333333333333. Then reply with the single word DONE. Do not call any other tool."
diff --git a/examples/acp-agent/tests/snapshots/subagent-list-agents/session.jsonl b/examples/acp-agent/tests/snapshots/subagent-list-agents/session.jsonl
index 06bd463ba8..44549b2e99 100644
--- a/examples/acp-agent/tests/snapshots/subagent-list-agents/session.jsonl
+++ b/examples/acp-agent/tests/snapshots/subagent-list-agents/session.jsonl
@@ -1,10 +1,10 @@
 {"type":"session","version":0,"id":"11111111-1111-4111-8111-111111111111","createdAt":1789000000000,"cwd":"{{cwd}}","delegationDepth":0}
-{"type":"agent/inbox/spliced","seq":0,"time":1785730454756,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Call the subagent tool once with run_in_background set to true, description 'Reply with CHILD_OK', and prompt 'Reply with exactly the word CHILD_OK and nothing else.'. Then reply with the single word STARTED. Do not call any other tool."}],"source":{"kind":"user"},"role":"user","id":"c2febfff-792d-4457-a944-933ff0de0570"}]}}
+{"type":"agent/inbox/spliced","seq":0,"time":1785730454756,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Call the subagent tool once with run_in_background set to true, description 'Reply with CHILD_OK', and prompt 'Reply with exactly the word CHILD_OK and nothing else.'. Then reply with the single word STARTED. Do not call any other tool."}],"source":{"kind":"user"},"role":"user","id":"356b3b62-c8b8-4d2a-84d7-7df1b6e4811e"}]}}
 {"type":"turn/start","seq":1,"time":1785821412725,"data":{"turn":1}}
 {"type":"agent/inbox/spliced","seq":2,"time":1785821412725,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}
 {"type":"step/start","seq":3,"time":1785730454783,"data":{"turn":1,"step":1}}
-{"type":"user/message","seq":4,"time":1785730454783,"data":{"content":[{"type":"text","text":"Call the subagent tool once with run_in_background set to true, description 'Reply with CHILD_OK', and prompt 'Reply with exactly the word CHILD_OK and nothing else.'. Then reply with the single word STARTED. Do not call any other tool."}],"source":{"kind":"user"},"role":"user","id":"c2febfff-792d-4457-a944-933ff0de0570"},"surfaceOp":"append"}
-{"type":"user/message","seq":5,"time":1785730454783,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}]},"role":"user","id":"c9d1f853-56bc-4082-ae08-00d4bcbb04a6"},"surfaceOp":"append"}
+{"type":"user/message","seq":4,"time":1785730454783,"data":{"content":[{"type":"text","text":"Call the subagent tool once with run_in_background set to true, description 'Reply with CHILD_OK', and prompt 'Reply with exactly the word CHILD_OK and nothing else.'. Then reply with the single word STARTED. Do not call any other tool."}],"source":{"kind":"user"},"role":"user","id":"356b3b62-c8b8-4d2a-84d7-7df1b6e4811e"},"surfaceOp":"append"}
+{"type":"user/message","seq":5,"time":1785730454783,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}]},"role":"user","id":"9be42fb0-f0d0-4ab9-a232-fb753f7db482"},"surfaceOp":"append"}
 {"type":"session/title","seq":6,"time":1785730454783,"data":{"title":"Call the subagent tool once","messageSeqs":[4],"source":{"kind":"fallback"}}}
 {"type":"request/header","seq":7,"time":1785730454784,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}}
 {"type":"request/context","seq":8,"time":1785730454784,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}}
@@ -13,9 +13,9 @@
 {"type":"assistant/chunk","seq":11,"time":1785531795632,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Reply with CHILD_OK\", \"prompt\": \"Reply with exactly the word CHILD_OK and nothing else.\", \"run_in_background\": true}"}}}}
 {"type":"assistant/chunk","seq":12,"time":1785531795632,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
 {"type":"assistant/chunk","seq":13,"time":1785730454793,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
-{"type":"assistant/message","seq":14,"time":1785730454793,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Reply with CHILD_OK\", \"prompt\": \"Reply with exactly the word CHILD_OK and nothing else.\", \"run_in_background\": true}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"4dda49d6-5d02-456f-ba95-68699662793d"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[9,10,11,12,13],"surfaceOp":"append"}
+{"type":"assistant/message","seq":14,"time":1785730454793,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Reply with CHILD_OK\", \"prompt\": \"Reply with exactly the word CHILD_OK and nothing else.\", \"run_in_background\": true}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"c8802574-4e43-4ee7-8648-5a132935b5dc"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[9,10,11,12,13],"surfaceOp":"append"}
 {"type":"tool/call","seq":15,"time":1785730454793,"data":{"turn":1,"step":1,"callId":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Reply with CHILD_OK\", \"prompt\": \"Reply with exactly the word CHILD_OK and nothing else.\", \"run_in_background\": true}"}}
-{"type":"tool/result","seq":16,"time":1785730454804,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_bg_start"},"content":[{"type":"tool-result","toolCallId":"call_bg_start","content":[{"type":"text","text":"started subagent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"7f1fbe79-98b7-410a-af4d-c40d52a366dc"}},"sourceEventSeqs":[15],"surfaceOp":"append"}
+{"type":"tool/result","seq":16,"time":1785730454804,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_bg_start"},"content":[{"type":"tool-result","toolCallId":"call_bg_start","content":[{"type":"text","text":"started subagent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"c83395ad-93c6-4899-9ae1-8d29f92d4dde"}},"sourceEventSeqs":[15],"surfaceOp":"append"}
 {"type":"step/end","seq":17,"time":1785730454804,"data":{"turn":1,"step":1}}
 {"type":"step/start","seq":18,"time":1785730454814,"data":{"turn":1,"step":2}}
 {"type":"assistant/chunk","seq":19,"time":1789000000017,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
@@ -23,39 +23,42 @@
 {"type":"assistant/chunk","seq":21,"time":1785531795656,"data":{"turn":1,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"STARTED"}}}}
 {"type":"assistant/chunk","seq":22,"time":1785531795656,"data":{"turn":1,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
 {"type":"assistant/chunk","seq":23,"time":1785730454820,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}
-{"type":"assistant/message","seq":24,"time":1785730454820,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"text","text":"STARTED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"2ee459fa-21f9-48c6-a42e-3c38eca1e4c9"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"}
+{"type":"assistant/message","seq":24,"time":1785730454820,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"text","text":"STARTED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"1fefe87b-4c3c-49b0-860c-8097193f9567"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"}
 {"type":"step/end","seq":25,"time":1785730454821,"data":{"turn":1,"step":2}}
 {"type":"turn/end","seq":26,"time":1785730454821,"data":{"turn":1,"reason":{"kind":"completed"}}}
-{"type":"agent/inbox/spliced","seq":27,"time":1785730454857,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Call list_agents once with scope set to descendants and observe the subagent you started. Then call interrupt_agent once with agent_id set to 33333333-3333-4333-8333-333333333333. Then reply with the single word DONE. Do not call any other tool."}],"source":{"kind":"user"},"role":"user","id":"3bcac2a3-e0db-465b-94b4-4e2761236475"}]}}
-{"type":"turn/start","seq":28,"time":1785821412840,"data":{"turn":2}}
-{"type":"agent/inbox/spliced","seq":29,"time":1785821412840,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}
-{"type":"step/start","seq":30,"time":1785730454863,"data":{"turn":2,"step":1}}
-{"type":"user/message","seq":31,"time":1785730454863,"data":{"content":[{"type":"text","text":"Call list_agents once with scope set to descendants and observe the subagent you started. Then call interrupt_agent once with agent_id set to 33333333-3333-4333-8333-333333333333. Then reply with the single word DONE. Do not call any other tool."}],"source":{"kind":"user"},"role":"user","id":"3bcac2a3-e0db-465b-94b4-4e2761236475"},"surfaceOp":"append"}
-{"type":"assistant/chunk","seq":32,"time":1785531795686,"data":{"turn":2,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}
-{"type":"assistant/chunk","seq":33,"time":1785536135065,"data":{"turn":2,"step":1,"chunk":{"type":"tool-call-delta","index":0,"id":"call_list","name":"list_agents","argumentsDelta":"{\"scope\":\"descendants\"}"}}}
-{"type":"assistant/chunk","seq":34,"time":1785730454867,"data":{"turn":2,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"call_list","name":"list_agents","arguments":"{\"scope\":\"descendants\"}"}}}}
-{"type":"assistant/chunk","seq":35,"time":1785730454867,"data":{"turn":2,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
-{"type":"assistant/chunk","seq":36,"time":1785730454867,"data":{"turn":2,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
-{"type":"assistant/message","seq":37,"time":1785730454867,"data":{"turn":2,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_list","name":"list_agents","arguments":"{\"scope\":\"descendants\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"199c7794-d782-4957-b7ed-69d094b9c0ef"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[32,33,34,35,36],"surfaceOp":"append"}
-{"type":"tool/call","seq":38,"time":1785730454867,"data":{"turn":2,"step":1,"callId":"call_list","name":"list_agents","arguments":"{\"scope\":\"descendants\"}"}}
-{"type":"tool/result","seq":39,"time":1785730454893,"data":{"turn":2,"step":1,"message":{"source":{"kind":"tool","callId":"call_list"},"content":[{"type":"tool-result","toolCallId":"call_list","content":[{"type":"text","text":"33333333-3333-4333-8333-333333333333 [complete] parent=11111111-1111-4111-8111-111111111111 depth=1 — Reply with CHILD_OK"}],"isError":false}],"role":"user","id":"61ceb650-3e32-413e-9d7e-b6ec8a351858"}},"sourceEventSeqs":[38],"surfaceOp":"append"}
-{"type":"step/end","seq":40,"time":1785730454893,"data":{"turn":2,"step":1}}
-{"type":"step/start","seq":41,"time":1785730454903,"data":{"turn":2,"step":2}}
-{"type":"assistant/chunk","seq":42,"time":1785531795715,"data":{"turn":2,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}
-{"type":"assistant/chunk","seq":43,"time":1785536135106,"data":{"turn":2,"step":2,"chunk":{"type":"tool-call-delta","index":0,"id":"call_interrupt","name":"interrupt_agent","argumentsDelta":"{\"agent_id\":\"33333333-3333-4333-8333-333333333333\"}"}}}
-{"type":"assistant/chunk","seq":44,"time":1785730454907,"data":{"turn":2,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"call_interrupt","name":"interrupt_agent","arguments":"{\"agent_id\":\"33333333-3333-4333-8333-333333333333\"}"}}}}
-{"type":"assistant/chunk","seq":45,"time":1785730454907,"data":{"turn":2,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
-{"type":"assistant/chunk","seq":46,"time":1785730454907,"data":{"turn":2,"step":2,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
-{"type":"assistant/message","seq":47,"time":1785730454907,"data":{"turn":2,"step":2,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_interrupt","name":"interrupt_agent","arguments":"{\"agent_id\":\"33333333-3333-4333-8333-333333333333\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"17c1a7c5-84f5-493e-adb5-6a65219e6ad6"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[42,43,44,45,46],"surfaceOp":"append"}
-{"type":"tool/call","seq":48,"time":1785730454907,"data":{"turn":2,"step":2,"callId":"call_interrupt","name":"interrupt_agent","arguments":"{\"agent_id\":\"33333333-3333-4333-8333-333333333333\"}"}}
-{"type":"tool/result","seq":49,"time":1785730454917,"data":{"turn":2,"step":2,"message":{"source":{"kind":"tool","callId":"call_interrupt"},"content":[{"type":"tool-result","toolCallId":"call_interrupt","content":[{"type":"text","text":"interrupt requested for agent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"bf93cbe0-1946-4aef-a4ce-431790026c6f"}},"sourceEventSeqs":[48],"surfaceOp":"append"}
-{"type":"step/end","seq":50,"time":1785730454917,"data":{"turn":2,"step":2}}
-{"type":"step/start","seq":51,"time":1785730454927,"data":{"turn":2,"step":3}}
-{"type":"assistant/chunk","seq":52,"time":1785531795715,"data":{"turn":2,"step":3,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
-{"type":"assistant/chunk","seq":53,"time":1785536135106,"data":{"turn":2,"step":3,"chunk":{"type":"text-delta","index":0,"text":"DONE"}}}
-{"type":"assistant/chunk","seq":54,"time":1785730454907,"data":{"turn":2,"step":3,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"DONE"}}}}
-{"type":"assistant/chunk","seq":55,"time":1785730454907,"data":{"turn":2,"step":3,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":2}}}}
-{"type":"assistant/chunk","seq":56,"time":1785730454907,"data":{"turn":2,"step":3,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}
-{"type":"assistant/message","seq":57,"time":1785730454907,"data":{"turn":2,"step":3,"message":{"role":"assistant","content":[{"type":"text","text":"DONE"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"e4c04cd6-9c23-4c99-a5f7-bac6fc141e95"},"usage":{"inputTokens":10,"outputTokens":2}},"sourceEventSeqs":[52,53,54,55,56],"surfaceOp":"append"}
-{"type":"step/end","seq":58,"time":1785730454938,"data":{"turn":2,"step":3}}
-{"type":"turn/end","seq":59,"time":1785730454938,"data":{"turn":2,"reason":{"kind":"completed"}}}
+{"type":"agent/inbox/spliced","seq":27,"time":1786011499211,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more."},{"type":"text","text":"Its closing message:"},{"type":"text","text":"CHILD_OK"}],"source":{"kind":"subagent-settled","form":"notice","summary":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more.","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"9275a12c-bf9a-48e2-b33b-4fc484e936cb"}]}}
+{"type":"turn/start","seq":28,"time":1786011499211,"data":{"turn":2}}
+{"type":"agent/inbox/spliced","seq":29,"time":1786011499211,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}
+{"type":"step/start","seq":30,"time":1786012470343,"data":{"turn":2,"step":1}}
+{"type":"user/message","seq":31,"time":1786012470343,"data":{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more."},{"type":"text","text":"Its closing message:"},{"type":"text","text":"CHILD_OK"}],"source":{"kind":"subagent-settled","form":"notice","summary":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more.","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"9275a12c-bf9a-48e2-b33b-4fc484e936cb"},"surfaceOp":"append"}
+{"type":"assistant/chunk","seq":32,"time":1786012470346,"data":{"turn":2,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
+{"type":"assistant/chunk","seq":33,"time":1786011499224,"data":{"turn":2,"step":1,"chunk":{"type":"text-delta","index":0,"text":"SUBAGENT_SETTLED_NOTED"}}}
+{"type":"assistant/chunk","seq":34,"time":1786011499224,"data":{"turn":2,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}}}}
+{"type":"assistant/chunk","seq":35,"time":1786011499224,"data":{"turn":2,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
+{"type":"assistant/chunk","seq":36,"time":1786011499224,"data":{"turn":2,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}
+{"type":"assistant/message","seq":37,"time":1786012470346,"data":{"turn":2,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"4620e8c0-dd13-4a2f-87dc-f4b66aa51219"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[32,33,34,35,36],"surfaceOp":"append"}
+{"type":"step/end","seq":38,"time":1786012470346,"data":{"turn":2,"step":1}}
+{"type":"turn/end","seq":39,"time":1786012470346,"data":{"turn":2,"reason":{"kind":"completed"}}}
+{"type":"agent/inbox/spliced","seq":40,"time":1786012470360,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Call list_agents once with scope set to descendants and observe the subagent you started. Then call interrupt_agent once with agent_id set to 33333333-3333-4333-8333-333333333333. Then reply with the single word DONE. Do not call any other tool."}],"source":{"kind":"user"},"role":"user","id":"7a2a86d0-80a3-4db5-822f-2d3fcbc16e11"}]}}
+{"type":"turn/start","seq":41,"time":1786011499224,"data":{"turn":3}}
+{"type":"agent/inbox/spliced","seq":42,"time":1786011499224,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}
+{"type":"step/start","seq":43,"time":1786011499234,"data":{"turn":3,"step":1}}
+{"type":"user/message","seq":44,"time":1786011499234,"data":{"content":[{"type":"text","text":"Call list_agents once with scope set to descendants and observe the subagent you started. Then call interrupt_agent once with agent_id set to 33333333-3333-4333-8333-333333333333. Then reply with the single word DONE. Do not call any other tool."}],"source":{"kind":"user"},"role":"user","id":"7a2a86d0-80a3-4db5-822f-2d3fcbc16e11"},"surfaceOp":"append"}
+{"type":"assistant/chunk","seq":45,"time":1786011499238,"data":{"turn":3,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}
+{"type":"assistant/chunk","seq":46,"time":1786011499238,"data":{"turn":3,"step":1,"chunk":{"type":"tool-call-delta","index":0,"id":"call_list","name":"list_agents","argumentsDelta":"{}"}}}
+{"type":"assistant/chunk","seq":47,"time":1785531795715,"data":{"turn":3,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"call_list","name":"list_agents","arguments":"{}"}}}}
+{"type":"assistant/chunk","seq":48,"time":1785536135106,"data":{"turn":3,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
+{"type":"assistant/chunk","seq":49,"time":1785730454907,"data":{"turn":3,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
+{"type":"assistant/message","seq":50,"time":1786011499238,"data":{"turn":3,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_list","name":"list_agents","arguments":"{}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"d3402e92-2f7e-4cd5-9537-ae9beedeecab"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[45,46,47,48,49],"surfaceOp":"append"}
+{"type":"tool/call","seq":51,"time":1786011499238,"data":{"turn":3,"step":1,"callId":"call_list","name":"list_agents","arguments":"{}"}}
+{"type":"tool/result","seq":52,"time":1786011499267,"data":{"turn":3,"step":1,"message":{"source":{"kind":"tool","callId":"call_list"},"content":[{"type":"tool-result","toolCallId":"call_list","content":[{"type":"text","text":"33333333-3333-4333-8333-333333333333 [ready] — Reply with CHILD_OK"}],"isError":false}],"role":"user","id":"8ae233de-8fde-48d7-a9d0-0d9a480a00d0"}},"sourceEventSeqs":[51],"surfaceOp":"append"}
+{"type":"step/end","seq":53,"time":1785730454908,"data":{"turn":3,"step":1}}
+{"type":"step/start","seq":54,"time":1786011499275,"data":{"turn":3,"step":2}}
+{"type":"assistant/chunk","seq":55,"time":1786011499279,"data":{"turn":3,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
+{"type":"assistant/chunk","seq":56,"time":1786011499279,"data":{"turn":3,"step":2,"chunk":{"type":"text-delta","index":0,"text":"DONE"}}}
+{"type":"assistant/chunk","seq":57,"time":1786011499279,"data":{"turn":3,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"DONE"}}}}
+{"type":"assistant/chunk","seq":58,"time":1786011499279,"data":{"turn":3,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":2}}}}
+{"type":"assistant/chunk","seq":59,"time":1786011499279,"data":{"turn":3,"step":2,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}
+{"type":"assistant/message","seq":60,"time":1786011499279,"data":{"turn":3,"step":2,"message":{"role":"assistant","content":[{"type":"text","text":"DONE"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"3ac0f29f-72ae-44fb-9414-974470095618"},"usage":{"inputTokens":10,"outputTokens":2}},"sourceEventSeqs":[55,56,57,58,59],"surfaceOp":"append"}
+{"type":"step/end","seq":61,"time":1786011499279,"data":{"turn":3,"step":2}}
+{"type":"turn/end","seq":62,"time":1786011499279,"data":{"turn":3,"reason":{"kind":"completed"}}}
diff --git a/examples/acp-agent/tests/snapshots/subagent-list-agents/stdout.expected.jsonl b/examples/acp-agent/tests/snapshots/subagent-list-agents/stdout.expected.jsonl
index 5a9b409384..4813c19f90 100644
--- a/examples/acp-agent/tests/snapshots/subagent-list-agents/stdout.expected.jsonl
+++ b/examples/acp-agent/tests/snapshots/subagent-list-agents/stdout.expected.jsonl
@@ -2,5 +2,6 @@
 {"jsonrpc":"2.0","id":2,"result":{"sessionId":"{{sessionId}}"}}
 {"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"STARTED"}}}}
 {"jsonrpc":"2.0","id":3,"result":{"stopReason":"end_turn"}}
+{"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}}}}
 {"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"DONE"}}}}
 {"jsonrpc":"2.0","id":4,"result":{"stopReason":"end_turn"}}
diff --git a/examples/acp-agent/tests/snapshots/subagent-list-agents/system-prompt.1.expected.md b/examples/acp-agent/tests/snapshots/subagent-list-agents/system-prompt.1.expected.md
new file mode 100644
index 0000000000..71f4ed05f8
--- /dev/null
+++ b/examples/acp-agent/tests/snapshots/subagent-list-agents/system-prompt.1.expected.md
@@ -0,0 +1,24 @@
+You are an AI agent powered by the DeepSeek Harness SDK.
+
+You are a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}. Your bash tool runs under a file sandbox — a `[sandbox: file access denied …]` result is policy, not a command bug.
+
+Verify your work by running the code or tests. Keep answers brief and factual.
+
+
+Use the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.
+
+Use the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.
+
+Use the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.
+
+Check the [exit code: N] marker on every bash result; investigate failures before moving on.
+
+Track every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.
+
+Use goal tools for one long-running completion objective in the current session. create_goal may infer goal intent from a direct human request in any language; do not create a goal for routine single-turn work. Call get_goal before update_goal and copy its exact goal_id and revision. After session resume or fork, an active goal is disarmed: when a human asks to continue or resume in any wording or language, use update_goal action resume to rearm it. Mark complete only when the objective is actually achieved. Mark blocked only after the same blocking condition persists for at least 3 consecutive goal rounds, and report that concrete condition in blocked_reason; difficulty, uncertainty, or useful remaining work is not blocked.
+
+Use the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.
+
+Use the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.
+
+Deliver your result with the report tool before you finish: call it once with a self-contained answer. The agent that started you shares your workspace but does not automatically receive your transcript, tool output, or reasoning, so a closing remark such as "done" leaves it nothing it can use. Report earlier as well whenever a partial finding changes what that agent should do next; reporting never ends your turn.
diff --git a/examples/acp-agent/tests/snapshots/subagent-list-agents/tool-schemas.1.expected.json b/examples/acp-agent/tests/snapshots/subagent-list-agents/tool-schemas.1.expected.json
index 7dd791cf27..253fbc4a32 100644
--- a/examples/acp-agent/tests/snapshots/subagent-list-agents/tool-schemas.1.expected.json
+++ b/examples/acp-agent/tests/snapshots/subagent-list-agents/tool-schemas.1.expected.json
@@ -133,7 +133,7 @@
     },
     {
       "name": "list_agents",
-      "description": "List your continuable background subagents by durable id and label. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and complete means it exists only in storage — a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
+      "description": "List your continuable background subagents by durable id and label. Use it to recall which ones you started, not to poll for completion — you are told when one finishes. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and ready means it exists only in storage — resumable, not terminal, and not a result waiting to be collected; a `send_message` starts a new turn on the same conversation, and a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -194,13 +194,13 @@
     },
     {
       "name": "report",
-      "description": "Report selected content to the agent that started you. Call this zero or more times for progress, findings, or a final answer. Reporting does not end your turn or finish your work, and only your direct parent receives it. A failed call may still have arrived, so do not blindly repeat it.",
+      "description": "Report selected content to the agent that started you. Call this once before you finish, with a self-contained final result, and earlier for progress or findings that change what that agent does next. That agent shares your workspace but does not automatically receive your transcript, tool output, or reasoning, so finishing your work is not itself a result. Reporting does not end your turn or finish your work, and only your direct parent receives it. A failed call may still have arrived, so do not blindly repeat it.",
       "parameters": {
         "type": "object",
         "properties": {
           "output": {
             "type": "string",
-            "description": "Self-contained content for your parent; it does not see your private work."
+            "description": "Actionable content for your parent; summarize conclusions and reference relevant shared paths."
           }
         },
         "required": [
@@ -247,7 +247,7 @@
     },
     {
       "name": "subagent",
-      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -261,7 +261,7 @@
           },
           "run_in_background": {
             "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
+            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."
           }
         },
         "required": [
@@ -272,7 +272,7 @@
     },
     {
       "name": "subagent_fork",
-      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -283,10 +283,6 @@
           "prompt": {
             "type": "string",
             "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."
-          },
-          "run_in_background": {
-            "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
           }
         },
         "required": [
diff --git a/examples/acp-agent/tests/snapshots/subagent-mixed/session.1.jsonl b/examples/acp-agent/tests/snapshots/subagent-mixed/session.1.jsonl
index e510929344..ca5b3abd3a 100644
--- a/examples/acp-agent/tests/snapshots/subagent-mixed/session.1.jsonl
+++ b/examples/acp-agent/tests/snapshots/subagent-mixed/session.1.jsonl
@@ -6,7 +6,7 @@
 {"type":"subagent/descriptor","seq":4,"time":1786357524752,"data":{"version":2,"mode":"one-shot","provider":"spawn","label":"Reply ALPHA only"}}
 {"type":"step/start","seq":5,"time":1786357524755,"data":{"turn":1,"step":1}}
 {"type":"user/message","seq":6,"time":1785730450187,"data":{"content":[{"type":"text","text":"Reply with exactly the word ALPHA and nothing else."}],"source":{"kind":"user"},"role":"user","id":"73ce401a-faaf-408a-879e-7485380d537d"},"surfaceOp":"append"}
-{"type":"user/message","seq":7,"time":1786357524755,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"56ad93fa-0cc0-4ccb-a1b4-258f4801c681"},"surfaceOp":"append"}
+{"type":"user/message","seq":7,"time":1786357524755,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"cb073ca7-b412-4509-99ef-4e1351e7e74d"},"surfaceOp":"append"}
 {"type":"session/title","seq":8,"time":1786357524755,"data":{"title":"Reply with exactly the word","messageSeqs":[6],"source":{"kind":"fallback"}}}
 {"type":"request/header","seq":9,"time":1785730450187,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}}
 {"type":"request/context","seq":10,"time":1785730450188,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}}
diff --git a/examples/acp-agent/tests/snapshots/subagent-mixed/session.2.jsonl b/examples/acp-agent/tests/snapshots/subagent-mixed/session.2.jsonl
index a5f4ab88a4..d846a2bcd8 100644
--- a/examples/acp-agent/tests/snapshots/subagent-mixed/session.2.jsonl
+++ b/examples/acp-agent/tests/snapshots/subagent-mixed/session.2.jsonl
@@ -27,7 +27,7 @@
 {"type":"subagent/descriptor","seq":41,"time":1786357524800,"data":{"version":2,"mode":"one-shot","provider":"fork","label":"Recall project codeword"}}
 {"type":"step/start","seq":42,"time":1786357524803,"data":{"turn":2,"step":1}}
 {"type":"user/message","seq":43,"time":1786357524803,"data":{"content":[{"type":"text","text":"What is the project codeword mentioned earlier in this conversation? Reply with exactly that one word and nothing else."}],"source":{"kind":"user"},"role":"user","id":"86e9f144-764f-460d-b72b-262cffe43d77"},"surfaceOp":"append"}
-{"type":"user/message","seq":44,"time":1786358036899,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"6ea5a774-b0da-47ff-84b7-226a4a207bbf"},"surfaceOp":"append"}
+{"type":"user/message","seq":44,"time":1786358036899,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"d5ebfdfe-de5f-4f3d-9772-7d6179fe8c3d"},"surfaceOp":"append"}
 {"type":"request/header","seq":45,"time":1786358036900,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"resume"}}
 {"type":"assistant/chunk","seq":46,"time":1783352148077,"data":{"turn":2,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"reasoning"}}}
 {"type":"reasoning-chunks","seq0":47,"time0":1783352148077,"data":{"turn":2,"step":1,"index":0,"dt":[0,0,29,0,0,0,35,0,0,0,0,26,29,31,0,30,0,0,27,1,27,0,1,0,0,31,1,0,0,1790157964],"texts":["The"," user"," is"," asking"," me"," to"," recall"," the"," project"," cod","ew","ord"," that"," was"," mentioned"," earlier"," in"," the"," conversation","."," I"," was"," told"," to"," remember"," it",":"," SA","FF","RON","."]}}
diff --git a/examples/acp-agent/tests/snapshots/subagent-multi/session.1.jsonl b/examples/acp-agent/tests/snapshots/subagent-multi/session.1.jsonl
index da79d6b23c..45fbf5e48b 100644
--- a/examples/acp-agent/tests/snapshots/subagent-multi/session.1.jsonl
+++ b/examples/acp-agent/tests/snapshots/subagent-multi/session.1.jsonl
@@ -6,7 +6,7 @@
 {"type":"subagent/descriptor","seq":4,"time":1786357521754,"data":{"version":2,"mode":"one-shot","provider":"spawn","label":"Return ALPHA only"}}
 {"type":"step/start","seq":5,"time":1786357521756,"data":{"turn":1,"step":1}}
 {"type":"user/message","seq":6,"time":1785730447828,"data":{"content":[{"type":"text","text":"Reply with exactly the word ALPHA and nothing else."}],"source":{"kind":"user"},"role":"user","id":"a287f842-f6f2-4a17-ab4c-820e41f498d5"},"surfaceOp":"append"}
-{"type":"user/message","seq":7,"time":1786357521756,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"f654b5a4-b4c0-4443-8eab-d84624d804f1"},"surfaceOp":"append"}
+{"type":"user/message","seq":7,"time":1786357521756,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"6f423049-484c-48d6-9cdd-db504d79f892"},"surfaceOp":"append"}
 {"type":"session/title","seq":8,"time":1786357521756,"data":{"title":"Reply with exactly the word","messageSeqs":[6],"source":{"kind":"fallback"}}}
 {"type":"request/header","seq":9,"time":1785730447828,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}}
 {"type":"request/context","seq":10,"time":1785730447828,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}}
diff --git a/examples/acp-agent/tests/snapshots/subagent-multi/session.2.jsonl b/examples/acp-agent/tests/snapshots/subagent-multi/session.2.jsonl
index f7326e37c8..7b390e04d4 100644
--- a/examples/acp-agent/tests/snapshots/subagent-multi/session.2.jsonl
+++ b/examples/acp-agent/tests/snapshots/subagent-multi/session.2.jsonl
@@ -6,7 +6,7 @@
 {"type":"subagent/descriptor","seq":4,"time":1786357521799,"data":{"version":2,"mode":"one-shot","provider":"spawn","label":"Return BETA only"}}
 {"type":"step/start","seq":5,"time":1786357521801,"data":{"turn":1,"step":1}}
 {"type":"user/message","seq":6,"time":1785730447881,"data":{"content":[{"type":"text","text":"Reply with exactly the word BETA and nothing else."}],"source":{"kind":"user"},"role":"user","id":"53f6419d-8ddc-4eee-8803-5b68411336f9"},"surfaceOp":"append"}
-{"type":"user/message","seq":7,"time":1786357521802,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"05843da4-4a5f-46fc-a00f-5257b2bd271d"},"surfaceOp":"append"}
+{"type":"user/message","seq":7,"time":1786357521802,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"5832009f-12a3-4f6e-85e5-257c06f4c716"},"surfaceOp":"append"}
 {"type":"session/title","seq":8,"time":1786357521802,"data":{"title":"Reply with exactly the word","messageSeqs":[6],"source":{"kind":"fallback"}}}
 {"type":"request/header","seq":9,"time":1785730447881,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}}
 {"type":"request/context","seq":10,"time":1785730447881,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}}
diff --git a/examples/acp-agent/tests/snapshots/subagent-parallel/session.1.jsonl b/examples/acp-agent/tests/snapshots/subagent-parallel/session.1.jsonl
index 3555ca83de..d5f1733ce0 100644
--- a/examples/acp-agent/tests/snapshots/subagent-parallel/session.1.jsonl
+++ b/examples/acp-agent/tests/snapshots/subagent-parallel/session.1.jsonl
@@ -1,18 +1,18 @@
 {"type":"session","version":0,"id":"bbbbbbbb-0000-4000-8000-000000000002","createdAt":1783352127000,"cwd":"{{cwd}}","parentSession":"aaaaaaaa-0000-4000-8000-000000000001","origin":"subagent","delegationDepth":1}
 {"type":"approval/policy","seq":0,"time":1786373947132,"data":{"policy":"never","source":"delegation"}}
-{"type":"agent/inbox/spliced","seq":1,"time":1786373947134,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Reply with exactly the word ALPHA and nothing else."}],"source":{"kind":"user"},"role":"user","id":"f5d60a48-78a3-4d75-91d7-478017516c5c"}]}}
+{"type":"agent/inbox/spliced","seq":1,"time":1786373947134,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Reply with exactly the word ALPHA and nothing else."}],"source":{"kind":"user"},"role":"user","id":"84b2bc8e-878b-4337-ae82-26c3074069f2"}]}}
 {"type":"turn/start","seq":2,"time":1786373947134,"data":{"turn":1}}
 {"type":"agent/inbox/spliced","seq":3,"time":1786373947134,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}
 {"type":"subagent/descriptor","seq":4,"time":1786373947165,"data":{"version":2,"mode":"one-shot","provider":"spawn","label":"Say the word ALPHA"}}
 {"type":"step/start","seq":5,"time":1786373947168,"data":{"turn":1,"step":1}}
-{"type":"user/message","seq":6,"time":1786338530759,"data":{"content":[{"type":"text","text":"Reply with exactly the word ALPHA and nothing else."}],"source":{"kind":"user"},"role":"user","id":"f5d60a48-78a3-4d75-91d7-478017516c5c"},"surfaceOp":"append"}
-{"type":"user/message","seq":7,"time":1786373947168,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"635fd4be-77c6-4891-90ba-2cafa163e166"},"surfaceOp":"append"}
+{"type":"user/message","seq":6,"time":1786338530759,"data":{"content":[{"type":"text","text":"Reply with exactly the word ALPHA and nothing else."}],"source":{"kind":"user"},"role":"user","id":"84b2bc8e-878b-4337-ae82-26c3074069f2"},"surfaceOp":"append"}
+{"type":"user/message","seq":7,"time":1786373947168,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"05fbd821-3adf-4877-b805-2cc5e6de3f9c"},"surfaceOp":"append"}
 {"type":"session/title","seq":8,"time":1786373947168,"data":{"title":"Reply with exactly the word","messageSeqs":[6],"source":{"kind":"fallback"}}}
 {"type":"request/header","seq":9,"time":1786338530759,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}}
 {"type":"request/context","seq":10,"time":1786338530759,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}}
 {"type":"assistant/chunk","seq":11,"time":1786338530769,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
 {"type":"assistant/chunk","seq":12,"time":1786338530769,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"ALPHA"}}}}
 {"type":"assistant/chunk","seq":13,"time":1786338530769,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}
-{"type":"assistant/message","seq":14,"time":1786338530769,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"ALPHA"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"4d13c6bf-dc50-4e57-91b0-59561b58986e"}},"sourceEventSeqs":[11,12,13],"surfaceOp":"append"}
+{"type":"assistant/message","seq":14,"time":1786338530769,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"ALPHA"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"224c83f1-4449-451d-8f8a-f2889c19068e"}},"sourceEventSeqs":[11,12,13],"surfaceOp":"append"}
 {"type":"step/end","seq":15,"time":1786338530769,"data":{"turn":1,"step":1}}
 {"type":"turn/end","seq":16,"time":1786338530769,"data":{"turn":1,"reason":{"kind":"completed"}}}
diff --git a/examples/acp-agent/tests/snapshots/subagent-parallel/session.2.jsonl b/examples/acp-agent/tests/snapshots/subagent-parallel/session.2.jsonl
index 6bb570bdaf..aea0d83108 100644
--- a/examples/acp-agent/tests/snapshots/subagent-parallel/session.2.jsonl
+++ b/examples/acp-agent/tests/snapshots/subagent-parallel/session.2.jsonl
@@ -1,18 +1,18 @@
 {"type":"session","version":0,"id":"cccccccc-0000-4000-8000-000000000003","createdAt":1783352127001,"cwd":"{{cwd}}","parentSession":"aaaaaaaa-0000-4000-8000-000000000001","origin":"subagent","delegationDepth":1}
 {"type":"approval/policy","seq":0,"time":1786373947132,"data":{"policy":"never","source":"delegation"}}
-{"type":"agent/inbox/spliced","seq":1,"time":1786373947133,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Reply with exactly the word ALPHA and nothing else."}],"source":{"kind":"user"},"role":"user","id":"afacd216-5e1b-46e7-afc4-f0eba9da1814"}]}}
+{"type":"agent/inbox/spliced","seq":1,"time":1786373947133,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Reply with exactly the word ALPHA and nothing else."}],"source":{"kind":"user"},"role":"user","id":"a3da85dc-6428-4a31-8653-b7327e518436"}]}}
 {"type":"turn/start","seq":2,"time":1786373947133,"data":{"turn":1}}
 {"type":"agent/inbox/spliced","seq":3,"time":1786373947134,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}
 {"type":"subagent/descriptor","seq":4,"time":1786373947170,"data":{"version":2,"mode":"one-shot","provider":"spawn","label":"Say the word ALPHA"}}
 {"type":"step/start","seq":5,"time":1786373947173,"data":{"turn":1,"step":1}}
-{"type":"user/message","seq":6,"time":1786338530749,"data":{"content":[{"type":"text","text":"Reply with exactly the word ALPHA and nothing else."}],"source":{"kind":"user"},"role":"user","id":"afacd216-5e1b-46e7-afc4-f0eba9da1814"},"surfaceOp":"append"}
-{"type":"user/message","seq":7,"time":1786373947174,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"dde02975-239c-422e-9b63-27d98f73346a"},"surfaceOp":"append"}
+{"type":"user/message","seq":6,"time":1786338530749,"data":{"content":[{"type":"text","text":"Reply with exactly the word ALPHA and nothing else."}],"source":{"kind":"user"},"role":"user","id":"a3da85dc-6428-4a31-8653-b7327e518436"},"surfaceOp":"append"}
+{"type":"user/message","seq":7,"time":1786373947174,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"670f5fc4-3eb6-4430-959a-08e8adf25036"},"surfaceOp":"append"}
 {"type":"session/title","seq":8,"time":1786373947174,"data":{"title":"Reply with exactly the word","messageSeqs":[6],"source":{"kind":"fallback"}}}
 {"type":"request/header","seq":9,"time":1786338530749,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}}
 {"type":"request/context","seq":10,"time":1786338530749,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}}
 {"type":"assistant/chunk","seq":11,"time":1786338530759,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
 {"type":"assistant/chunk","seq":12,"time":1786338530759,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"ALPHA"}}}}
 {"type":"assistant/chunk","seq":13,"time":1786338530759,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}
-{"type":"assistant/message","seq":14,"time":1786338530759,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"ALPHA"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"11e93a53-9112-4e80-9447-de974373ca94"}},"sourceEventSeqs":[11,12,13],"surfaceOp":"append"}
+{"type":"assistant/message","seq":14,"time":1786338530759,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"ALPHA"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"ac2dc91f-7dfa-45f6-88e0-21a2e762c3dd"}},"sourceEventSeqs":[11,12,13],"surfaceOp":"append"}
 {"type":"step/end","seq":15,"time":1786338530760,"data":{"turn":1,"step":1}}
 {"type":"turn/end","seq":16,"time":1786338530760,"data":{"turn":1,"reason":{"kind":"completed"}}}
diff --git a/examples/acp-agent/tests/snapshots/subagent-report/input.json b/examples/acp-agent/tests/snapshots/subagent-report/input.json
index b15e0e4d71..15ca4b53eb 100644
--- a/examples/acp-agent/tests/snapshots/subagent-report/input.json
+++ b/examples/acp-agent/tests/snapshots/subagent-report/input.json
@@ -14,6 +14,13 @@
     {
       "op": "waitForSubagentTurnEnd"
     },
+    {
+      "op": "waitForTurnStart",
+      "minimumTurn": 2
+    },
+    {
+      "op": "waitForTurnEnd"
+    },
     {
       "op": "promptAndWaitForAgentMessage",
       "text": "Repeat back, verbatim, the exact output the background subagent reported to you. Reply with only that text. Do not use any tools.",
diff --git a/examples/acp-agent/tests/snapshots/subagent-report/session.1.jsonl b/examples/acp-agent/tests/snapshots/subagent-report/session.1.jsonl
index e32d1bdee0..b83358f93d 100644
--- a/examples/acp-agent/tests/snapshots/subagent-report/session.1.jsonl
+++ b/examples/acp-agent/tests/snapshots/subagent-report/session.1.jsonl
@@ -1,14 +1,14 @@
 {"type":"session","version":0,"id":"33333333-3333-4333-8333-333333333333","createdAt":1789000001000,"cwd":"{{cwd}}","parentSession":"11111111-1111-4111-8111-111111111111","origin":"subagent","delegationDepth":1}
 {"type":"subagent/descriptor","seq":0,"time":1785594881508,"data":{"version":2,"mode":"continuable","provider":"spawn","label":"Report a finding","agentProvider":"deepseek-official","agentModel":"deepseek-v4-flash"}}
 {"type":"session/end-seed","seq":1,"time":1785594881508,"data":{}}
-{"type":"approval/policy","seq":2,"time":1786357530605,"data":{"policy":"never","source":"delegation"}}
-{"type":"agent/inbox/spliced","seq":3,"time":1786357530605,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Call the report tool once with output exactly CHILD_REPORT_OK, then stop."}],"source":{"kind":"user"},"role":"user","id":"9045ac78-393a-4f24-b20d-8999286dd6ce"}]}}
-{"type":"turn/start","seq":4,"time":1786357530605,"data":{"turn":1}}
-{"type":"agent/inbox/spliced","seq":5,"time":1786357530605,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}
-{"type":"step/start","seq":6,"time":1786357530633,"data":{"turn":1,"step":1}}
+{"type":"approval/policy","seq":2,"time":1786374158805,"data":{"policy":"never","source":"delegation"}}
+{"type":"agent/inbox/spliced","seq":3,"time":1786374158806,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Call the report tool once with output exactly CHILD_REPORT_OK, then stop."}],"source":{"kind":"user"},"role":"user","id":"9045ac78-393a-4f24-b20d-8999286dd6ce"}]}}
+{"type":"turn/start","seq":4,"time":1786374158806,"data":{"turn":1}}
+{"type":"agent/inbox/spliced","seq":5,"time":1786374158806,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}
+{"type":"step/start","seq":6,"time":1786374158840,"data":{"turn":1,"step":1}}
 {"type":"user/message","seq":7,"time":1785730453639,"data":{"content":[{"type":"text","text":"Call the report tool once with output exactly CHILD_REPORT_OK, then stop."}],"source":{"kind":"user"},"role":"user","id":"9045ac78-393a-4f24-b20d-8999286dd6ce"},"surfaceOp":"append"}
-{"type":"user/message","seq":8,"time":1786357530633,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"1677b901-cce7-461a-8b2a-7f119dd9d845"},"surfaceOp":"append"}
-{"type":"session/title","seq":9,"time":1786357530633,"data":{"title":"Call the report tool once","messageSeqs":[7],"source":{"kind":"fallback"}}}
+{"type":"user/message","seq":8,"time":1786374158840,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"0f107d71-9b56-4ad8-b6f1-d93cb4c82105"},"surfaceOp":"append"}
+{"type":"session/title","seq":9,"time":1786374158840,"data":{"title":"Call the report tool once","messageSeqs":[7],"source":{"kind":"fallback"}}}
 {"type":"request/header","seq":10,"time":1785730453639,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}}
 {"type":"request/context","seq":11,"time":1785730453639,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}}
 {"type":"assistant/chunk","seq":12,"time":1785594881546,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}
@@ -18,7 +18,7 @@
 {"type":"assistant/chunk","seq":16,"time":1785730453647,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
 {"type":"assistant/message","seq":17,"time":1785730453647,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_report_1","name":"report","arguments":"{\"output\": \"CHILD_REPORT_OK\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"c9e50afb-b732-41ab-b0fc-8e98948ad9ec"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[12,13,14,15,16],"surfaceOp":"append"}
 {"type":"tool/call","seq":18,"time":1785730453647,"data":{"turn":1,"step":1,"callId":"call_report_1","name":"report","arguments":"{\"output\": \"CHILD_REPORT_OK\"}"}}
-{"type":"tool/result","seq":19,"time":1785730453654,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_report_1"},"content":[{"type":"tool-result","toolCallId":"call_report_1","content":[{"type":"text","text":"report accepted by the agent that started you as message 1f4b61e2-6c6d-4db8-836b-ac5760c5e484"}],"isError":false}],"role":"user","id":"cee5f084-bfab-423d-b8bc-1b1b7d88d4fa"}},"sourceEventSeqs":[18],"surfaceOp":"append"}
+{"type":"tool/result","seq":19,"time":1785730453654,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_report_1"},"content":[{"type":"tool-result","toolCallId":"call_report_1","content":[{"type":"text","text":"report accepted by the agent that started you as message 87627538-d804-4d36-bb10-4768b6fcfb65"}],"isError":false}],"role":"user","id":"22f25be2-d3f9-4558-b3ea-db22fa900aa3"}},"sourceEventSeqs":[18],"surfaceOp":"append"}
 {"type":"step/end","seq":20,"time":1785730453654,"data":{"turn":1,"step":1}}
 {"type":"step/start","seq":21,"time":1785730453664,"data":{"turn":1,"step":2}}
 {"type":"assistant/chunk","seq":22,"time":1785594881567,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
diff --git a/examples/acp-agent/tests/snapshots/subagent-report/session.jsonl b/examples/acp-agent/tests/snapshots/subagent-report/session.jsonl
index eb36af2399..f8b5deef29 100644
--- a/examples/acp-agent/tests/snapshots/subagent-report/session.jsonl
+++ b/examples/acp-agent/tests/snapshots/subagent-report/session.jsonl
@@ -1,10 +1,10 @@
 {"type":"session","version":0,"id":"11111111-1111-4111-8111-111111111111","createdAt":1789000000000,"cwd":"{{cwd}}","delegationDepth":0}
-{"type":"agent/inbox/spliced","seq":0,"time":1785730453561,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Follow these steps exactly, then stop. 1. Call the subagent tool once with run_in_background set to true, description 'Report a finding', and prompt 'Call the report tool once with output exactly CHILD_REPORT_OK, then stop.'. 2. Reply with the single word STARTED. Do not use the bash tool."}],"source":{"kind":"user"},"role":"user","id":"5cf78378-e004-4fd5-af4f-cef3b7e190ad"}]}}
+{"type":"agent/inbox/spliced","seq":0,"time":1785730453561,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Follow these steps exactly, then stop. 1. Call the subagent tool once with run_in_background set to true, description 'Report a finding', and prompt 'Call the report tool once with output exactly CHILD_REPORT_OK, then stop.'. 2. Reply with the single word STARTED. Do not use the bash tool."}],"source":{"kind":"user"},"role":"user","id":"b765ae32-73e2-4625-81ba-01095f8c83d0"}]}}
 {"type":"turn/start","seq":1,"time":1785821411429,"data":{"turn":1}}
 {"type":"agent/inbox/spliced","seq":2,"time":1785821411429,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}
 {"type":"step/start","seq":3,"time":1785730453591,"data":{"turn":1,"step":1}}
-{"type":"user/message","seq":4,"time":1785730453591,"data":{"content":[{"type":"text","text":"Follow these steps exactly, then stop. 1. Call the subagent tool once with run_in_background set to true, description 'Report a finding', and prompt 'Call the report tool once with output exactly CHILD_REPORT_OK, then stop.'. 2. Reply with the single word STARTED. Do not use the bash tool."}],"source":{"kind":"user"},"role":"user","id":"5cf78378-e004-4fd5-af4f-cef3b7e190ad"},"surfaceOp":"append"}
-{"type":"user/message","seq":5,"time":1785730453592,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}]},"role":"user","id":"d1a851a3-604f-4a42-8e5f-4e480857a3b4"},"surfaceOp":"append"}
+{"type":"user/message","seq":4,"time":1785730453591,"data":{"content":[{"type":"text","text":"Follow these steps exactly, then stop. 1. Call the subagent tool once with run_in_background set to true, description 'Report a finding', and prompt 'Call the report tool once with output exactly CHILD_REPORT_OK, then stop.'. 2. Reply with the single word STARTED. Do not use the bash tool."}],"source":{"kind":"user"},"role":"user","id":"b765ae32-73e2-4625-81ba-01095f8c83d0"},"surfaceOp":"append"}
+{"type":"user/message","seq":5,"time":1785730453592,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}]},"role":"user","id":"1f4f5888-2068-4df0-904f-12ffb4aa3321"},"surfaceOp":"append"}
 {"type":"session/title","seq":6,"time":1785730453592,"data":{"title":"Follow these steps exactly, then","messageSeqs":[4],"source":{"kind":"fallback"}}}
 {"type":"request/header","seq":7,"time":1785730453592,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}}
 {"type":"request/context","seq":8,"time":1785730453593,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}}
@@ -13,9 +13,9 @@
 {"type":"assistant/chunk","seq":11,"time":1785501592851,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Report a finding\", \"prompt\": \"Call the report tool once with output exactly CHILD_REPORT_OK, then stop.\", \"run_in_background\": true}"}}}}
 {"type":"assistant/chunk","seq":12,"time":1785501592851,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
 {"type":"assistant/chunk","seq":13,"time":1785730453601,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
-{"type":"assistant/message","seq":14,"time":1785730453602,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Report a finding\", \"prompt\": \"Call the report tool once with output exactly CHILD_REPORT_OK, then stop.\", \"run_in_background\": true}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"664c39ff-9dac-4bb3-a151-e18a7863d15a"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[9,10,11,12,13],"surfaceOp":"append"}
+{"type":"assistant/message","seq":14,"time":1785730453602,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Report a finding\", \"prompt\": \"Call the report tool once with output exactly CHILD_REPORT_OK, then stop.\", \"run_in_background\": true}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"97b897d5-0d01-4a6c-ad0c-4776c61c9c68"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[9,10,11,12,13],"surfaceOp":"append"}
 {"type":"tool/call","seq":15,"time":1785730453602,"data":{"turn":1,"step":1,"callId":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Report a finding\", \"prompt\": \"Call the report tool once with output exactly CHILD_REPORT_OK, then stop.\", \"run_in_background\": true}"}}
-{"type":"tool/result","seq":16,"time":1785730453613,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_bg_start"},"content":[{"type":"tool-result","toolCallId":"call_bg_start","content":[{"type":"text","text":"started subagent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"d7927ad2-29db-4de8-ab4d-59a4ebcddd72"}},"sourceEventSeqs":[15],"surfaceOp":"append"}
+{"type":"tool/result","seq":16,"time":1785730453613,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_bg_start"},"content":[{"type":"tool-result","toolCallId":"call_bg_start","content":[{"type":"text","text":"started subagent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"91fb94ce-cf3e-47ed-ab20-8f46cf4aec55"}},"sourceEventSeqs":[15],"surfaceOp":"append"}
 {"type":"step/end","seq":17,"time":1785730453613,"data":{"turn":1,"step":1}}
 {"type":"step/start","seq":18,"time":1785730453623,"data":{"turn":1,"step":2}}
 {"type":"assistant/chunk","seq":19,"time":1789000000018,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
@@ -23,22 +23,35 @@
 {"type":"assistant/chunk","seq":21,"time":1785501592877,"data":{"turn":1,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"STARTED"}}}}
 {"type":"assistant/chunk","seq":22,"time":1785501592877,"data":{"turn":1,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
 {"type":"assistant/chunk","seq":23,"time":1785730453628,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}
-{"type":"assistant/message","seq":24,"time":1785730453628,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"text","text":"STARTED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"571faad7-adbd-480c-922a-1499e1329ead"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"}
+{"type":"assistant/message","seq":24,"time":1785730453628,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"text","text":"STARTED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"b1b1cf78-11a8-4440-b9f9-2096d15e7884"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"}
 {"type":"step/end","seq":25,"time":1785730453628,"data":{"turn":1,"step":2}}
 {"type":"turn/end","seq":26,"time":1785730453629,"data":{"turn":1,"reason":{"kind":"completed"}}}
-{"type":"agent/inbox/spliced","seq":27,"time":1785730453654,"data":{"target":"next-step","start":0,"inserted":[{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 reported:"},{"type":"text","text":"CHILD_REPORT_OK"}],"source":{"kind":"subagent-report","form":"relay","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"824dc60a-f9d7-48ea-a0d4-6d56df83bd4f"}]}}
-{"type":"agent/inbox/spliced","seq":28,"time":1785730453673,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Repeat back, verbatim, the exact output the background subagent reported to you. Reply with only that text. Do not use any tools."}],"source":{"kind":"user"},"role":"user","id":"43f17984-22c3-48b9-911e-923a2f68dce0"}]}}
-{"type":"turn/start","seq":29,"time":1785821411548,"data":{"turn":2}}
-{"type":"agent/inbox/spliced","seq":30,"time":1785730453673,"data":{"target":"next-step","start":0,"removedCount":1,"inserted":[]}}
-{"type":"agent/inbox/spliced","seq":31,"time":1785821411548,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}
-{"type":"step/start","seq":32,"time":1785730453683,"data":{"turn":2,"step":1}}
-{"type":"user/message","seq":33,"time":1785730453683,"data":{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 reported:"},{"type":"text","text":"CHILD_REPORT_OK"}],"source":{"kind":"subagent-report","form":"relay","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"824dc60a-f9d7-48ea-a0d4-6d56df83bd4f"},"surfaceOp":"append"}
-{"type":"user/message","seq":34,"time":1785730453683,"data":{"content":[{"type":"text","text":"Repeat back, verbatim, the exact output the background subagent reported to you. Reply with only that text. Do not use any tools."}],"source":{"kind":"user"},"role":"user","id":"43f17984-22c3-48b9-911e-923a2f68dce0"},"surfaceOp":"append"}
-{"type":"assistant/chunk","seq":35,"time":1785730453687,"data":{"turn":2,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
-{"type":"assistant/chunk","seq":36,"time":1785730453687,"data":{"turn":2,"step":1,"chunk":{"type":"text-delta","index":0,"text":"CHILD_REPORT_OK"}}}
-{"type":"assistant/chunk","seq":37,"time":1785730453687,"data":{"turn":2,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"CHILD_REPORT_OK"}}}}
-{"type":"assistant/chunk","seq":38,"time":1785730453687,"data":{"turn":2,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
-{"type":"assistant/chunk","seq":39,"time":1785730453687,"data":{"turn":2,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}
-{"type":"assistant/message","seq":40,"time":1785730453687,"data":{"turn":2,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"CHILD_REPORT_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"77bba235-d2b6-4a32-9ba3-ebb69d9b0654"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[35,36,37,38,39],"surfaceOp":"append"}
-{"type":"step/end","seq":41,"time":1785730453687,"data":{"turn":2,"step":1}}
-{"type":"turn/end","seq":42,"time":1785730453687,"data":{"turn":2,"reason":{"kind":"completed"}}}
+{"type":"agent/inbox/spliced","seq":27,"time":1786008565468,"data":{"target":"next-step","start":0,"inserted":[{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 reported:"},{"type":"text","text":"CHILD_REPORT_OK"}],"source":{"kind":"subagent-report","form":"relay","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"ec024a7a-5506-4ebf-a9d8-82ce01dc88b4"}]}}
+{"type":"agent/inbox/spliced","seq":28,"time":1786012334062,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more."},{"type":"text","text":"Its closing message:"},{"type":"text","text":"Reported."}],"source":{"kind":"subagent-settled","form":"notice","summary":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more.","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"08101cc1-abde-49ca-9745-1d075a3911b5"}]}}
+{"type":"turn/start","seq":29,"time":1786012334062,"data":{"turn":2}}
+{"type":"agent/inbox/spliced","seq":30,"time":1786012334062,"data":{"target":"next-step","start":0,"removedCount":1,"inserted":[]}}
+{"type":"agent/inbox/spliced","seq":31,"time":1786012334062,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}
+{"type":"step/start","seq":32,"time":1786376144953,"data":{"turn":2,"step":1}}
+{"type":"user/message","seq":33,"time":1786376144953,"data":{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 reported:"},{"type":"text","text":"CHILD_REPORT_OK"}],"source":{"kind":"subagent-report","form":"relay","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"ec024a7a-5506-4ebf-a9d8-82ce01dc88b4"},"surfaceOp":"append"}
+{"type":"user/message","seq":34,"time":1786012334068,"data":{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more."},{"type":"text","text":"Its closing message:"},{"type":"text","text":"Reported."}],"source":{"kind":"subagent-settled","form":"notice","summary":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more.","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"08101cc1-abde-49ca-9745-1d075a3911b5"},"surfaceOp":"append"}
+{"type":"assistant/chunk","seq":35,"time":1786376144959,"data":{"turn":2,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
+{"type":"assistant/chunk","seq":36,"time":1786012334073,"data":{"turn":2,"step":1,"chunk":{"type":"text-delta","index":0,"text":"SUBAGENT_SETTLED_NOTED"}}}
+{"type":"assistant/chunk","seq":37,"time":1786012334073,"data":{"turn":2,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}}}}
+{"type":"assistant/chunk","seq":38,"time":1786012334073,"data":{"turn":2,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
+{"type":"assistant/chunk","seq":39,"time":1786012334073,"data":{"turn":2,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}
+{"type":"assistant/message","seq":40,"time":1786376144959,"data":{"turn":2,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"c6fecd6e-033f-4be8-98ee-cc0733b18c83"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[35,36,37,38,39],"surfaceOp":"append"}
+{"type":"step/end","seq":41,"time":1786376144960,"data":{"turn":2,"step":1}}
+{"type":"turn/end","seq":42,"time":1786376144960,"data":{"turn":2,"reason":{"kind":"completed"}}}
+{"type":"agent/inbox/spliced","seq":43,"time":1786376145165,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Repeat back, verbatim, the exact output the background subagent reported to you. Reply with only that text. Do not use any tools."}],"source":{"kind":"user"},"role":"user","id":"6c0b0e51-4ad9-4c4b-bbbc-508973862b77"}]}}
+{"type":"turn/start","seq":44,"time":1786012334073,"data":{"turn":3}}
+{"type":"agent/inbox/spliced","seq":45,"time":1786012334073,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}
+{"type":"step/start","seq":46,"time":1786012334082,"data":{"turn":3,"step":1}}
+{"type":"user/message","seq":47,"time":1786012334083,"data":{"content":[{"type":"text","text":"Repeat back, verbatim, the exact output the background subagent reported to you. Reply with only that text. Do not use any tools."}],"source":{"kind":"user"},"role":"user","id":"6c0b0e51-4ad9-4c4b-bbbc-508973862b77"},"surfaceOp":"append"}
+{"type":"assistant/chunk","seq":48,"time":1786012334087,"data":{"turn":3,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
+{"type":"assistant/chunk","seq":49,"time":1786012334087,"data":{"turn":3,"step":1,"chunk":{"type":"text-delta","index":0,"text":"CHILD_REPORT_OK"}}}
+{"type":"assistant/chunk","seq":50,"time":1786012334087,"data":{"turn":3,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"CHILD_REPORT_OK"}}}}
+{"type":"assistant/chunk","seq":51,"time":1786012334087,"data":{"turn":3,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
+{"type":"assistant/chunk","seq":52,"time":1786012334087,"data":{"turn":3,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}
+{"type":"assistant/message","seq":53,"time":1786012334087,"data":{"turn":3,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"CHILD_REPORT_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"98d44266-6695-482b-910c-0e1e570fe7a6"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[48,49,50,51,52],"surfaceOp":"append"}
+{"type":"step/end","seq":54,"time":1786012334087,"data":{"turn":3,"step":1}}
+{"type":"turn/end","seq":55,"time":1786012334087,"data":{"turn":3,"reason":{"kind":"completed"}}}
diff --git a/examples/acp-agent/tests/snapshots/subagent-report/stdout.expected.jsonl b/examples/acp-agent/tests/snapshots/subagent-report/stdout.expected.jsonl
index 09cd3e4a68..785ee28128 100644
--- a/examples/acp-agent/tests/snapshots/subagent-report/stdout.expected.jsonl
+++ b/examples/acp-agent/tests/snapshots/subagent-report/stdout.expected.jsonl
@@ -2,5 +2,6 @@
 {"jsonrpc":"2.0","id":2,"result":{"sessionId":"{{sessionId}}"}}
 {"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"STARTED"}}}}
 {"jsonrpc":"2.0","id":3,"result":{"stopReason":"end_turn"}}
+{"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}}}}
 {"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"CHILD_REPORT_OK"}}}}
 {"jsonrpc":"2.0","id":4,"result":{"stopReason":"end_turn"}}
diff --git a/examples/acp-agent/tests/snapshots/subagent-report/system-prompt.1.expected.md b/examples/acp-agent/tests/snapshots/subagent-report/system-prompt.1.expected.md
new file mode 100644
index 0000000000..71f4ed05f8
--- /dev/null
+++ b/examples/acp-agent/tests/snapshots/subagent-report/system-prompt.1.expected.md
@@ -0,0 +1,24 @@
+You are an AI agent powered by the DeepSeek Harness SDK.
+
+You are a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}. Your bash tool runs under a file sandbox — a `[sandbox: file access denied …]` result is policy, not a command bug.
+
+Verify your work by running the code or tests. Keep answers brief and factual.
+
+
+Use the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.
+
+Use the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.
+
+Use the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.
+
+Check the [exit code: N] marker on every bash result; investigate failures before moving on.
+
+Track every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.
+
+Use goal tools for one long-running completion objective in the current session. create_goal may infer goal intent from a direct human request in any language; do not create a goal for routine single-turn work. Call get_goal before update_goal and copy its exact goal_id and revision. After session resume or fork, an active goal is disarmed: when a human asks to continue or resume in any wording or language, use update_goal action resume to rearm it. Mark complete only when the objective is actually achieved. Mark blocked only after the same blocking condition persists for at least 3 consecutive goal rounds, and report that concrete condition in blocked_reason; difficulty, uncertainty, or useful remaining work is not blocked.
+
+Use the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.
+
+Use the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.
+
+Deliver your result with the report tool before you finish: call it once with a self-contained answer. The agent that started you shares your workspace but does not automatically receive your transcript, tool output, or reasoning, so a closing remark such as "done" leaves it nothing it can use. Report earlier as well whenever a partial finding changes what that agent should do next; reporting never ends your turn.
diff --git a/examples/acp-agent/tests/snapshots/subagent-report/tool-schemas.1.expected.json b/examples/acp-agent/tests/snapshots/subagent-report/tool-schemas.1.expected.json
index 7dd791cf27..253fbc4a32 100644
--- a/examples/acp-agent/tests/snapshots/subagent-report/tool-schemas.1.expected.json
+++ b/examples/acp-agent/tests/snapshots/subagent-report/tool-schemas.1.expected.json
@@ -133,7 +133,7 @@
     },
     {
       "name": "list_agents",
-      "description": "List your continuable background subagents by durable id and label. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and complete means it exists only in storage — a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
+      "description": "List your continuable background subagents by durable id and label. Use it to recall which ones you started, not to poll for completion — you are told when one finishes. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and ready means it exists only in storage — resumable, not terminal, and not a result waiting to be collected; a `send_message` starts a new turn on the same conversation, and a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -194,13 +194,13 @@
     },
     {
       "name": "report",
-      "description": "Report selected content to the agent that started you. Call this zero or more times for progress, findings, or a final answer. Reporting does not end your turn or finish your work, and only your direct parent receives it. A failed call may still have arrived, so do not blindly repeat it.",
+      "description": "Report selected content to the agent that started you. Call this once before you finish, with a self-contained final result, and earlier for progress or findings that change what that agent does next. That agent shares your workspace but does not automatically receive your transcript, tool output, or reasoning, so finishing your work is not itself a result. Reporting does not end your turn or finish your work, and only your direct parent receives it. A failed call may still have arrived, so do not blindly repeat it.",
       "parameters": {
         "type": "object",
         "properties": {
           "output": {
             "type": "string",
-            "description": "Self-contained content for your parent; it does not see your private work."
+            "description": "Actionable content for your parent; summarize conclusions and reference relevant shared paths."
           }
         },
         "required": [
@@ -247,7 +247,7 @@
     },
     {
       "name": "subagent",
-      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -261,7 +261,7 @@
           },
           "run_in_background": {
             "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
+            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."
           }
         },
         "required": [
@@ -272,7 +272,7 @@
     },
     {
       "name": "subagent_fork",
-      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -283,10 +283,6 @@
           "prompt": {
             "type": "string",
             "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."
-          },
-          "run_in_background": {
-            "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
           }
         },
         "required": [
diff --git a/examples/acp-agent/tests/snapshots/text-turn/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/text-turn/tool-schemas.expected.json
index e0b7a41bcf..a2629a3153 100644
--- a/examples/acp-agent/tests/snapshots/text-turn/tool-schemas.expected.json
+++ b/examples/acp-agent/tests/snapshots/text-turn/tool-schemas.expected.json
@@ -133,7 +133,7 @@
     },
     {
       "name": "list_agents",
-      "description": "List your continuable background subagents by durable id and label. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and complete means it exists only in storage — a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
+      "description": "List your continuable background subagents by durable id and label. Use it to recall which ones you started, not to poll for completion — you are told when one finishes. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and ready means it exists only in storage — resumable, not terminal, and not a result waiting to be collected; a `send_message` starts a new turn on the same conversation, and a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -231,7 +231,7 @@
     },
     {
       "name": "subagent",
-      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -245,7 +245,7 @@
           },
           "run_in_background": {
             "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
+            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."
           }
         },
         "required": [
@@ -256,7 +256,7 @@
     },
     {
       "name": "subagent_fork",
-      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -267,10 +267,6 @@
           "prompt": {
             "type": "string",
             "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."
-          },
-          "run_in_background": {
-            "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
           }
         },
         "required": [
diff --git a/examples/acp-agent/tests/snapshots/web-fetch/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/web-fetch/tool-schemas.expected.json
index 97c51af14b..f11d079c63 100644
--- a/examples/acp-agent/tests/snapshots/web-fetch/tool-schemas.expected.json
+++ b/examples/acp-agent/tests/snapshots/web-fetch/tool-schemas.expected.json
@@ -133,7 +133,7 @@
     },
     {
       "name": "list_agents",
-      "description": "List your continuable background subagents by durable id and label. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and complete means it exists only in storage — a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
+      "description": "List your continuable background subagents by durable id and label. Use it to recall which ones you started, not to poll for completion — you are told when one finishes. Status comes from the live registry: running means the agent is working right now, idle means it is loaded but between turns (it may be waiting on agents it started), and ready means it exists only in storage — resumable, not terminal, and not a result waiting to be collected; a `send_message` starts a new turn on the same conversation, and a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery promise — `send_message` performs the authoritative check and may still fail. Children that could not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent session id and depth. You may use `send_message` only for depth-1 entries; deeper entries are candidates for `interrupt_agent` only.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -231,7 +231,7 @@
     },
     {
       "name": "subagent",
-      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -245,7 +245,7 @@
           },
           "run_in_background": {
             "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
+            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."
           }
         },
         "required": [
@@ -256,7 +256,7 @@
     },
     {
       "name": "subagent_fork",
-      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.",
+      "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.",
       "parameters": {
         "type": "object",
         "properties": {
@@ -267,10 +267,6 @@
           "prompt": {
             "type": "string",
             "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."
-          },
-          "run_in_background": {
-            "type": "boolean",
-            "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."
           }
         },
         "required": [
diff --git a/examples/headless-agent/cordis.yml b/examples/headless-agent/cordis.yml
index b6f1b774f2..61a0190142 100644
--- a/examples/headless-agent/cordis.yml
+++ b/examples/headless-agent/cordis.yml
@@ -118,12 +118,16 @@
     backgroundMode: continuable
     maxDepth: 1
 
+# Fork stays one-shot because a continuable child's `report` tool and prompt
+# section precede the inherited history a fork reuses; `run_in_background` is off
+# because this example mounts no task service. See .agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md.
 - id: tool-subagent-fork
   name: '@deepseek-ai/dsh-tool-subagent'
   config:
     provider: fork
     toolName: subagent_fork
-    backgroundMode: continuable
+    backgroundMode: one-shot
+    enableRunInBackground: false
     maxDepth: 1
 
 # The worker-thread workflow engine fans a model-written JavaScript script's
diff --git a/examples/headless-agent/subagent-settlement.cordis.snapshot.yml b/examples/headless-agent/subagent-settlement.cordis.snapshot.yml
new file mode 100644
index 0000000000..4812d676a2
--- /dev/null
+++ b/examples/headless-agent/subagent-settlement.cordis.snapshot.yml
@@ -0,0 +1,21 @@
+# Keyless assembled-app coverage for continuable child settlement delivery. The
+# replay child deliberately never calls report; the parent can reach its final
+# answer only if the continuation manager places the child's closing message in
+# the parent turn without list_agents, send_message, or a Task collector.
+
+- id: base
+  name: '@deepseek-ai/cordis-plugin-include'
+  config:
+    path: ./cordis.yml
+    patches:
+      - id: llm-deepseek
+        name: '@deepseek-ai/dsh-llm-deepseek'
+        disabled: true
+      - insert:
+          - id: llm-replay
+            name: '@deepseek-ai/dsh-llm-replay'
+
+# Prevent platform scheduling from choosing a streamed-chunk interleave. The
+# fence releases only after the real manager notice enters the parent inbox.
+- id: settlement-fence
+  name: './tests/fixtures/subagent-settlement-fence.ts'
diff --git a/examples/headless-agent/tests/fixtures/subagent-settlement-fence.ts b/examples/headless-agent/tests/fixtures/subagent-settlement-fence.ts
new file mode 100644
index 0000000000..c2c8a87885
--- /dev/null
+++ b/examples/headless-agent/tests/fixtures/subagent-settlement-fence.ts
@@ -0,0 +1,43 @@
+/**
+ * Loader fixture that holds the parent's second step until settlement delivery.
+ * @module subagent-settlement-fence
+ */
+
+import type { Context } from '@deepseek-ai/cordis'
+import type {} from '@deepseek-ai/dsh-agent-loop'
+import type {} from '@deepseek-ai/dsh-subagent'
+
+/** Fixture plugin name. */
+export const name = 'subagent-settlement-fence'
+
+/**
+ * Fence the parent's post-spawn request behind admission of the manager notice.
+ *
+ * This pins content order, not step placement: the held pre-step runs after its
+ * own `Inbox.claim()`, so the notice lands after step 2's claim and is claimed at
+ * step 3 because the child's settlement pipeline is strictly longer than the
+ * parent's claim path, not because a barrier forces it.
+ * @param ctx - assembled headless-agent context.
+ */
+export function apply(ctx: Context): void {
+  const delivered = Promise.withResolvers<undefined>()
+  let hasDelivered = false
+
+  ctx.effect(() => {
+    const disposeInbox = ctx.root.on('agent/inbox/inserted', ({ agent, message }) => {
+      if (agent.session.header.parentSession !== undefined || message.source.kind !== 'subagent-settled') return
+      hasDelivered = true
+      delivered.resolve(undefined)
+    })
+    const disposeStep = ctx.root.on('agent/pre-step', async ({ agent, turn, step }, next) => {
+      if (agent.session.header.parentSession === undefined && turn === 1 && step === 2 && !hasDelivered) {
+        await delivered.promise
+      }
+      return next()
+    })
+    return () => {
+      disposeStep()
+      disposeInbox()
+    }
+  }, 'subagent-settlement-fence.listeners')
+}
diff --git a/examples/headless-agent/tests/headless.snapshot.ts b/examples/headless-agent/tests/headless.snapshot.ts
index 5fc337527e..61c1711566 100644
--- a/examples/headless-agent/tests/headless.snapshot.ts
+++ b/examples/headless-agent/tests/headless.snapshot.ts
@@ -45,6 +45,8 @@ const credentialsConfigPath = fileURLToPath(new URL('../credentials.cordis.snaps
 const invalidCredentialScenarioDir = join(snapshotsDir, 'invalid-credential')
 const ralphScenarioDir = join(snapshotsDir, 'ralph-loop')
 const ralphConfigPath = fileURLToPath(new URL('../ralph.cordis.snapshot.yml', import.meta.url))
+const settlementScenarioDir = join(snapshotsDir, 'subagent-settlement')
+const settlementConfigPath = fileURLToPath(new URL('../subagent-settlement.cordis.snapshot.yml', import.meta.url))
 const startupFailureConfigPath = fileURLToPath(new URL('./fixtures/startup-activation-error/cordis.yml', import.meta.url))
 const startupFailureExpected = join(snapshotsDir, 'startup-activation-error', 'stderr.expected.txt')
 const binScript = fileURLToPath(new URL('./fixtures/headless-driver.ts', import.meta.url))
@@ -777,6 +779,77 @@ describe('headless stream-json snapshots', () => {
     expect(normalized).toBe(await readFile(streamExpected, 'utf8'))
   }, LOADER_SMOKE_TEST_TIMEOUT_MS)
 
+  it('delivers a continuable child result without parent polling', async () => {
+    const parentReplay = join(settlementScenarioDir, 'parent.replay.jsonl')
+    const parentOverride = join(settlementScenarioDir, 'parent.override.json')
+    const childReplay = join(settlementScenarioDir, 'child.replay.jsonl')
+    const childExpected = join(settlementScenarioDir, 'child.expected.jsonl')
+    const streamExpected = join(settlementScenarioDir, 'stream-json.expected.jsonl')
+    const task = 'Start one continuable background subagent and answer from its completion notice. Do not call list_agents, send_message, task_output, or task_list.'
+    let runCwd = ''
+    const result = await runLoaderSmoke({
+      label: 'continuable settlement headless stream-json snapshot',
+      tempDirPrefix: 'headless-snapshot-subagent-settlement-',
+      binScript,
+      libBinScript: binScript,
+      configPath: settlementConfigPath,
+      binArgs: [settlementConfigPath, task],
+      tsconfigPath,
+      env: {
+        // The override fully supplies the parent script; the child fixture
+        // remains separate so replay binds it to the fresh child Session.
+        DSH_SNAPSHOT_FILE: parentReplay,
+        DSH_SNAPSHOT_OVERRIDE: parentOverride,
+        DSH_SNAPSHOT_CHILD_FILES: childReplay,
+        NODE_OPTIONS: [process.env.NODE_OPTIONS, '--disable-warning=ExperimentalWarning'].filter(Boolean).join(' '),
+      },
+      prepare: (cwd) => { runCwd = cwd },
+      inspect: async (cwd) => {
+        const logs = await persistedLogs(cwd)
+        expect(logs).toHaveLength(2)
+        const parent = logs.find(log => typeof log.header.parentSession !== 'string')
+        const child = logs.find(log => typeof log.header.parentSession === 'string')
+        if (parent === undefined || child === undefined) throw new Error('missing persisted parent or child log')
+
+        const parentRecords = parseJsonl(parent.content)
+        const calls = parentRecords.filter(record => record.type === 'tool/call')
+        expect(calls.map(record => (record.data as JsonObject | undefined)?.name)).toEqual(['subagent'])
+        const callArguments = (calls[0]?.data as JsonObject | undefined)?.arguments
+        if (typeof callArguments !== 'string') throw new Error('subagent call did not persist its arguments')
+        expect(JSON.parse(callArguments)).toMatchObject({ run_in_background: true })
+
+        const notices = parentRecords.flatMap((record) => {
+          if (record.type !== 'agent/inbox/spliced') return []
+          const inserted = (record.data as JsonObject | undefined)?.inserted
+          if (!Array.isArray(inserted)) return []
+          return (inserted as JsonObject[]).filter((message) => {
+            const source = message.source as JsonObject | undefined
+            return source?.kind === 'subagent-settled'
+          })
+        })
+        expect(notices).toHaveLength(1)
+        expect(JSON.stringify(notices[0])).toContain('CHILD_RESULT')
+
+        const context = contextFromLogs([parent.content, child.content])
+        const normalizedChild = scrubRequestHeaders(normalizeSessionLog(child.content, context))
+        if (refreshing) await writeFile(childExpected, normalizedChild)
+        expect(normalizedChild).toBe(await readFile(childExpected, 'utf8'))
+        expect(normalizedChild).toContain('CHILD_RESULT')
+        expect(normalizedChild).not.toContain('"name":"report"')
+      },
+    })
+
+    expect(result.stderr).toBe('')
+    const records = parseJsonl(result.stdout)
+    expect(records.at(-1)).toMatchObject({
+      type: 'result',
+      output: 'PARENT_RECEIVED_CHILD_RESULT',
+    })
+    const normalized = normalizeHeadlessStream(result.stdout, runCwd)
+    if (refreshing) await writeFile(streamExpected, normalized)
+    expect(normalized).toBe(await readFile(streamExpected, 'utf8'))
+  }, LOADER_SMOKE_TEST_TIMEOUT_MS)
+
   it('replays persistent PTY tools through the one-shot app', async () => {
     const input = JSON.parse(await readFile(join(ptyScenarioDir, 'input.json'), 'utf8')) as {
       steps?: { op?: unknown; text?: unknown }[]
diff --git a/examples/headless-agent/tests/snapshots/advanced-toolchain/session.1.jsonl b/examples/headless-agent/tests/snapshots/advanced-toolchain/session.1.jsonl
index 8a1c23140b..753bfc5406 100644
--- a/examples/headless-agent/tests/snapshots/advanced-toolchain/session.1.jsonl
+++ b/examples/headless-agent/tests/snapshots/advanced-toolchain/session.1.jsonl
@@ -1,19 +1,19 @@
 {"type":"session","version":0,"id":"22222222-2222-4222-8222-222222222222","createdAt":1783950001000,"cwd":"{{cwd}}","parentSession":"11111111-1111-4111-8111-111111111111","origin":"subagent","delegationDepth":1}
-{"type":"agent/inbox/spliced","seq":0,"time":1785498583877,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Reply with exactly DIRECT_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"1c70f3f7-2e85-4808-8376-03d4d3bee6e6"}]}}
+{"type":"agent/inbox/spliced","seq":0,"time":1785498583877,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Reply with exactly DIRECT_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"a5c63729-a1da-40df-9a7c-2fcae0861916"}]}}
 {"type":"turn/start","seq":1,"time":1785821454445,"data":{"turn":1}}
 {"type":"agent/inbox/spliced","seq":2,"time":1785821454445,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}
 {"type":"subagent/descriptor","seq":3,"time":1785821454466,"data":{"version":2,"mode":"one-shot","provider":"spawn","label":"Check direct child"}}
 {"type":"step/start","seq":4,"time":1785730501506,"data":{"turn":1,"step":1}}
-{"type":"user/message","seq":5,"time":1785730501506,"data":{"content":[{"type":"text","text":"Reply with exactly DIRECT_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"1c70f3f7-2e85-4808-8376-03d4d3bee6e6"},"surfaceOp":"append"}
-{"type":"user/message","seq":6,"time":1786358103673,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"c9a305d4-add2-453e-8789-4e5c127725f7"},"surfaceOp":"append"}
-{"type":"session/title","seq":7,"time":1786358103673,"data":{"title":"Reply with exactly DIRECT_CHILD_OK and","messageSeqs":[5],"source":{"kind":"fallback"}}}
-{"type":"request/header","seq":8,"time":1785498583897,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}.\n\nVerify your work by running the code or tests. Keep answers brief and factual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.\n\n## Writing code for run_code\n\nPass `run_code` the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped). Inside the program:\n\n- Call tools as `await tools.name(args)` — quoted access for exotic names: `tools[\"my-tool\"](args)`. Every call resolves to the tool's typed canonical JSON value. Tool arguments must be lossless JSON.\n- A FAILED tool call rejects with `ToolCallError`, whose `toolName` identifies the failed tool and whose `message` is human-readable — `try/catch` it to handle and continue.\n- Independent read-only calls MAY overlap under `Promise.all` (safe calls run concurrently; mutating calls run alone, in submission order). Sequence dependent work with `await`.\n- Emit results with `return` and/or `console.log(...)`. ONLY what you print or return comes back to you — intermediate tool results never enter the conversation, so extract just what you need.\n\nThe available tools:\n\n```ts\ntype JsonValue = null | boolean | number | string | JsonValue[] | { [key: string]: JsonValue }\n\ninterface ToolArgsMap {\n  /** Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under <mode> mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`. */\n  bash: {\n    /** The bash command to execute. */\n    command: string;\n    /** Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\". */\n    description: string;\n    /** Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry. */\n    timeoutMs?: number;\n    /** Working directory for this command. Defaults to the session workspace; a relative path is resolved against it. */\n    workdir?: string;\n    /** Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies. */\n    run_in_background?: boolean;\n  } & Record<string, JsonValue>;\n  /** Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc. */\n  cordis_inspect: {\n    /** Limit the report to one section. Omit for all sections. */\n    what?: \"services\" | \"plugins\" | \"tools\" | \"temporary\" | \"api\" | \"events\";\n    /** Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\". */\n    name?: string;\n  } & Record<string, JsonValue>;\n  /** Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement a normal local, project, or repository Plugin through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:<id>]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime. */\n  cordis_mount: {\n    /** JavaScript body returning a temporary Plugin; evaluated now and saved nowhere. */\n    code: string;\n  } & Record<string, JsonValue>;\n  /** Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins. */\n  cordis_unmount: {\n    /** The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart. */\n    id: string;\n  } & Record<string, JsonValue>;\n  /** Edit an existing UTF-8 text file by replacing literal text. */\n  edit: {\n    /** Path to edit, resolved by the filesystem backend. */\n    file_path: string;\n    /** Literal text to replace. Must match exactly. */\n    old_string: string;\n    /** Literal replacement text. Use an empty string to delete the match. */\n    new_string: string;\n    /** Replace all matches. Defaults to false; when false, old_string must appear exactly once. */\n    replace_all?: boolean;\n  } & Record<string, JsonValue>;\n  /** Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op. */\n  interrupt_agent: {\n    /** The agent id of the running agent to interrupt. */\n    agent_id: string;\n  } & Record<string, JsonValue>;\n  /** Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools. */\n  ralph: {\n    /** The immutable completion objective for every fresh Ralph round. */\n    objective: string;\n    /** Optional positive safe-integer round cap, bounded by the deployment ceiling. */\n    maxRounds?: number;\n  } & Record<string, JsonValue>;\n  /** Read a UTF-8 text file and return line-numbered content. */\n  read: {\n    /** Path to read, resolved by the filesystem backend. */\n    file_path: string;\n    /** 1-based first line to return. Defaults to 1. */\n    offset?: number;\n    /** Maximum number of lines to return. Defaults to 2000. */\n    limit?: number;\n  } & Record<string, JsonValue>;\n  /** Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered. */\n  send_message: {\n    /** The subagent id returned when the background subagent was started. */\n    subagent_id: string;\n    /** The message to deliver to the subagent. */\n    message: string;\n  } & Record<string, JsonValue>;\n  /** Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill. */\n  skill: {\n    /** The exact skill name from the available skills list. */\n    name: string;\n  } & Record<string, JsonValue>;\n  /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */\n  subagent: {\n    /** A short (3-5 word) description of the delegated task, for display. */\n    description: string;\n    /** The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs. */\n    prompt: string;\n    /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */\n    run_in_background?: boolean;\n  } & Record<string, JsonValue>;\n  /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */\n  subagent_fork: {\n    /** A short (3-5 word) description of the delegated task, for display. */\n    description: string;\n    /** The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new. */\n    prompt: string;\n    /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */\n    run_in_background?: boolean;\n  } & Record<string, JsonValue>;\n  /** Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops. */\n  task_kill: {\n    /** Task id returned by the tool that started the background work. */\n    task_id: string;\n    /** Optional short reason, recorded in the log and forwarded to the task. */\n    reason?: string;\n  } & Record<string, JsonValue>;\n  /** List your background tasks (running and finished) with their ids, kinds, and statuses. */\n  task_list: Record<string, JsonValue>;\n  /** Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap. */\n  task_output: {\n    /** Task id returned by the tool that started the background work. */\n    task_id: string;\n    /** Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive. */\n    wait?: boolean;\n    /** Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum. */\n    timeout_ms?: number;\n  } & Record<string, JsonValue>;\n  /** Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished). */\n  todo_write: {\n    /** The COMPLETE task list, replacing any previous list. */\n    todos: ({\n      /** What the task is — a short imperative line. */\n      content: string;\n      /** pending (not started) | in_progress (now) | completed (done). */\n      status: \"pending\" | \"in_progress\" | \"completed\";\n    })[];\n  } & Record<string, JsonValue>;\n  /** Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn. The workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return <value>` — the value must be JSON-serializable and is this tool's result. Script-body hooks: - `agent(prompt, opts?): Promise<any>` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly. - `pipeline(items, ...stages): Promise<any[]>` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages. - `parallel(thunks): Promise<any[]>` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`. - `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim. Misused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`. Constraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes. */\n  workflow: {\n    /** The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return <json-value>`). */\n    script: string;\n    /** The workflow identity block (plain JSON — never code). */\n    meta: {\n      /** Short kebab-case workflow name. */\n      name: string;\n      /** One-line description of what the workflow does. */\n      description: string;\n      /** Optional guidance on when this workflow applies. */\n      whenToUse?: string;\n      /** Optional phase declarations matched by phase() calls. */\n      phases?: ({\n        /** The phase title phase() calls match by exact string. */\n        title: string;\n        /** Optional one-line description of the phase. */\n        detail?: string;\n        /** Optional provider override this phase is expected to use. */\n        provider?: string;\n        /** Optional model override this phase is expected to use. */\n        model?: string;\n      } & Record<string, JsonValue>)[];\n    } & Record<string, JsonValue>;\n    /** Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}). */\n    args?: Record<string, JsonValue>;\n  } & Record<string, JsonValue>;\n  /** Create or fully replace a UTF-8 text file. */\n  write: {\n    /** Path to write, resolved by the filesystem backend. */\n    file_path: string;\n    /** Full UTF-8 text content to write. */\n    content: string;\n  } & Record<string, JsonValue>;\n}\n\ninterface ToolOutputMap {\n  bash: {\n    kind: \"background\";\n    taskId: string;\n  } | {\n    kind: \"foreground\";\n    exitCode: number | null;\n    signal: string | null;\n    timedOut: boolean;\n    aborted: boolean;\n    timeoutMs: number;\n    stdout: {\n      text: string;\n      truncated: boolean;\n      spillPath?: string;\n    };\n    stderr: {\n      text: string;\n      truncated: boolean;\n      spillPath?: string;\n    };\n    sandbox?: {\n      mode: string;\n      denied: boolean;\n      enforcement?: string;\n      runnerFailed?: boolean;\n    };\n  };\n  cordis_inspect: string;\n  cordis_mount: {\n    id: string;\n    pluginName: string;\n    state: \"pending\" | \"loading\" | \"active\" | \"failed\" | \"disposed\" | \"unloading\";\n    provides: string[];\n    waitingFor: string[];\n  };\n  cordis_unmount: {\n    id: string;\n    pluginName: string;\n  };\n  edit: {\n    path: string;\n    before: string;\n    after: string;\n  };\n  interrupt_agent: {\n    accepted: boolean;\n  };\n  ralph: {\n    runId: string;\n    agentsStarted: number;\n    result: JsonValue;\n  };\n  read: {\n    path: string;\n    offset: number;\n    lines: {\n      number: number;\n      text: string;\n    }[];\n    totalLines: number;\n  };\n  send_message: {\n    messageId: string;\n  };\n  skill: {\n    name: string;\n    provider: string;\n    resourceBase?: {\n      kind: \"directory\";\n      path: string;\n    } | {\n      kind: \"url\";\n      url: string;\n    } | {\n      kind: \"opaque\";\n      description: string;\n    };\n    content: string;\n  };\n  subagent: {\n    kind: \"background\";\n    taskId: string;\n  } | {\n    kind: \"continuable\";\n    subagentId: string;\n  } | {\n    kind: \"foreground\";\n    runId: string;\n    output: JsonValue[];\n  };\n  subagent_fork: {\n    kind: \"background\";\n    taskId: string;\n  } | {\n    kind: \"continuable\";\n    subagentId: string;\n  } | {\n    kind: \"foreground\";\n    runId: string;\n    output: JsonValue[];\n  };\n  task_kill: {\n    outcome: \"cancellation-requested\" | \"already-finished\";\n    task: {\n      id: string;\n      kind: string;\n      label: string;\n      status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n      detail?: string;\n      startedAt: number;\n      finishedAt?: number;\n    };\n  };\n  task_list: ({\n    id: string;\n    kind: string;\n    label: string;\n    status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n    detail?: string;\n    startedAt: number;\n    finishedAt?: number;\n  })[];\n  task_output: {\n    text: string;\n    task: {\n      id: string;\n      kind: string;\n      label: string;\n      status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n      detail?: string;\n      startedAt: number;\n      finishedAt?: number;\n    };\n  };\n  todo_write: {\n    todos: ({\n      content: string;\n      status: \"pending\" | \"in_progress\" | \"completed\";\n    })[];\n    counts: {\n      pending: number;\n      inProgress: number;\n      completed: number;\n    };\n  };\n  workflow: {\n    runId: string;\n    agentsStarted: number;\n    result: JsonValue;\n  };\n  write: {\n    path: string;\n    operation: \"create\" | \"update\";\n    before: string | null;\n    after: string;\n  };\n}\n\ntype ToolName = keyof ToolOutputMap\n\ndeclare class ToolCallError extends Error {\n  readonly name: \"ToolCallError\";\n  readonly toolName: ToolName;\n}\n\ndeclare const tools: {\n  [K in ToolName]: (args: ToolArgsMap[K]) => Promise<ToolOutputMap[K]>;\n}\n```","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under <mode> mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_inspect","description":"Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc.","parameters":{"type":"object","properties":{"what":{"type":"string","description":"Limit the report to one section. Omit for all sections.","enum":["services","plugins","tools","temporary","api","events"]},"name":{"type":"string","description":"Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\"."}}}},{"name":"cordis_mount","description":"Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement a normal local, project, or repository Plugin through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:<id>]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"JavaScript body returning a temporary Plugin; evaluated now and saved nowhere."}},"required":["code"]}},{"name":"cordis_unmount","description":"Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins.","parameters":{"type":"object","properties":{"id":{"type":"string","description":"The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart."}},"required":["id"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"run_code","description":"Execute a TypeScript program against the available tools. Write the BODY of an async function (erasable syntax only; top-level `await` and `return` work) and call tools as `await tools.name(args)` per the declarations in the system prompt. Only what you print or return comes back — curate it.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"The program: the body of an async TypeScript function."},"description":{"type":"string","description":"Clear, concise description of what this program does in active voice, 5-10 words (shown in the UI). Examples: \"Count TODO markers across packages\"; \"Read failing test and its fixture\"; \"Rename config key in every cordis.yml\"."}},"required":["code","description"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return <value>` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise<any>` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise<any[]>` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise<any[]>` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return <json-value>`)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}}
+{"type":"user/message","seq":5,"time":1785730501506,"data":{"content":[{"type":"text","text":"Reply with exactly DIRECT_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"a5c63729-a1da-40df-9a7c-2fcae0861916"},"surfaceOp":"append"}
+{"type":"user/message","seq":6,"time":1786373992181,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"4901536e-b7bd-4a4d-9f6f-192ac1d07ab6"},"surfaceOp":"append"}
+{"type":"session/title","seq":7,"time":1786373992181,"data":{"title":"Reply with exactly DIRECT_CHILD_OK and","messageSeqs":[5],"source":{"kind":"fallback"}}}
+{"type":"request/header","seq":8,"time":1785498583897,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}.\n\nVerify your work by running the code or tests. Keep answers brief and factual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.\n\n## Writing code for run_code\n\nPass `run_code` the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped). Inside the program:\n\n- Call tools as `await tools.name(args)` — quoted access for exotic names: `tools[\"my-tool\"](args)`. Every call resolves to the tool's typed canonical JSON value. Tool arguments must be lossless JSON.\n- A FAILED tool call rejects with `ToolCallError`, whose `toolName` identifies the failed tool and whose `message` is human-readable — `try/catch` it to handle and continue.\n- Independent read-only calls MAY overlap under `Promise.all` (safe calls run concurrently; mutating calls run alone, in submission order). Sequence dependent work with `await`.\n- Emit results with `return` and/or `console.log(...)`. ONLY what you print or return comes back to you — intermediate tool results never enter the conversation, so extract just what you need.\n\nThe available tools:\n\n```ts\ntype JsonValue = null | boolean | number | string | JsonValue[] | { [key: string]: JsonValue }\n\ninterface ToolArgsMap {\n  /** Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under <mode> mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`. */\n  bash: {\n    /** The bash command to execute. */\n    command: string;\n    /** Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\". */\n    description: string;\n    /** Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry. */\n    timeoutMs?: number;\n    /** Working directory for this command. Defaults to the session workspace; a relative path is resolved against it. */\n    workdir?: string;\n    /** Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies. */\n    run_in_background?: boolean;\n  } & Record<string, JsonValue>;\n  /** Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc. */\n  cordis_inspect: {\n    /** Limit the report to one section. Omit for all sections. */\n    what?: \"services\" | \"plugins\" | \"tools\" | \"temporary\" | \"api\" | \"events\";\n    /** Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\". */\n    name?: string;\n  } & Record<string, JsonValue>;\n  /** Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:<id>]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime. */\n  cordis_mount: {\n    /** JavaScript body returning a temporary Plugin; evaluated now and saved nowhere. */\n    code: string;\n  } & Record<string, JsonValue>;\n  /** Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins. */\n  cordis_unmount: {\n    /** The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart. */\n    id: string;\n  } & Record<string, JsonValue>;\n  /** Edit an existing UTF-8 text file by replacing literal text. */\n  edit: {\n    /** Path to edit, resolved by the filesystem backend. */\n    file_path: string;\n    /** Literal text to replace. Must match exactly. */\n    old_string: string;\n    /** Literal replacement text. Use an empty string to delete the match. */\n    new_string: string;\n    /** Replace all matches. Defaults to false; when false, old_string must appear exactly once. */\n    replace_all?: boolean;\n  } & Record<string, JsonValue>;\n  /** Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op. */\n  interrupt_agent: {\n    /** The agent id of the running agent to interrupt. */\n    agent_id: string;\n  } & Record<string, JsonValue>;\n  /** Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools. */\n  ralph: {\n    /** The immutable completion objective for every fresh Ralph round. */\n    objective: string;\n    /** Optional positive safe-integer round cap, bounded by the deployment ceiling. */\n    maxRounds?: number;\n  } & Record<string, JsonValue>;\n  /** Read a UTF-8 text file and return line-numbered content. */\n  read: {\n    /** Path to read, resolved by the filesystem backend. */\n    file_path: string;\n    /** 1-based first line to return. Defaults to 1. */\n    offset?: number;\n    /** Maximum number of lines to return. Defaults to 2000. */\n    limit?: number;\n  } & Record<string, JsonValue>;\n  /** Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered. */\n  send_message: {\n    /** The subagent id returned when the background subagent was started. */\n    subagent_id: string;\n    /** The message to deliver to the subagent. */\n    message: string;\n  } & Record<string, JsonValue>;\n  /** Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill. */\n  skill: {\n    /** The exact skill name from the available skills list. */\n    name: string;\n  } & Record<string, JsonValue>;\n  /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work. */\n  subagent: {\n    /** A short (3-5 word) description of the delegated task, for display. */\n    description: string;\n    /** The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs. */\n    prompt: string;\n    /** Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message. */\n    run_in_background?: boolean;\n  } & Record<string, JsonValue>;\n  /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. */\n  subagent_fork: {\n    /** A short (3-5 word) description of the delegated task, for display. */\n    description: string;\n    /** The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new. */\n    prompt: string;\n  } & Record<string, JsonValue>;\n  /** Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops. */\n  task_kill: {\n    /** Task id returned by the tool that started the background work. */\n    task_id: string;\n    /** Optional short reason, recorded in the log and forwarded to the task. */\n    reason?: string;\n  } & Record<string, JsonValue>;\n  /** List your background tasks (running and finished) with their ids, kinds, and statuses. */\n  task_list: Record<string, JsonValue>;\n  /** Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap. */\n  task_output: {\n    /** Task id returned by the tool that started the background work. */\n    task_id: string;\n    /** Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive. */\n    wait?: boolean;\n    /** Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum. */\n    timeout_ms?: number;\n  } & Record<string, JsonValue>;\n  /** Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished). */\n  todo_write: {\n    /** The COMPLETE task list, replacing any previous list. */\n    todos: ({\n      /** What the task is — a short imperative line. */\n      content: string;\n      /** pending (not started) | in_progress (now) | completed (done). */\n      status: \"pending\" | \"in_progress\" | \"completed\";\n    })[];\n  } & Record<string, JsonValue>;\n  /** Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn. The workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return <value>` — the value must be JSON-serializable and is this tool's result. Script-body hooks: - `agent(prompt, opts?): Promise<any>` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly. - `pipeline(items, ...stages): Promise<any[]>` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages. - `parallel(thunks): Promise<any[]>` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`. - `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim. Misused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`. Constraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes. */\n  workflow: {\n    /** The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return <json-value>`). */\n    script: string;\n    /** The workflow identity block (plain JSON — never code). */\n    meta: {\n      /** Short kebab-case workflow name. */\n      name: string;\n      /** One-line description of what the workflow does. */\n      description: string;\n      /** Optional guidance on when this workflow applies. */\n      whenToUse?: string;\n      /** Optional phase declarations matched by phase() calls. */\n      phases?: ({\n        /** The phase title phase() calls match by exact string. */\n        title: string;\n        /** Optional one-line description of the phase. */\n        detail?: string;\n        /** Optional provider override this phase is expected to use. */\n        provider?: string;\n        /** Optional model override this phase is expected to use. */\n        model?: string;\n      } & Record<string, JsonValue>)[];\n    } & Record<string, JsonValue>;\n    /** Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}). */\n    args?: Record<string, JsonValue>;\n  } & Record<string, JsonValue>;\n  /** Create or fully replace a UTF-8 text file. */\n  write: {\n    /** Path to write, resolved by the filesystem backend. */\n    file_path: string;\n    /** Full UTF-8 text content to write. */\n    content: string;\n  } & Record<string, JsonValue>;\n}\n\ninterface ToolOutputMap {\n  bash: {\n    kind: \"background\";\n    taskId: string;\n  } | {\n    kind: \"foreground\";\n    exitCode: number | null;\n    signal: string | null;\n    timedOut: boolean;\n    aborted: boolean;\n    timeoutMs: number;\n    stdout: {\n      text: string;\n      truncated: boolean;\n      spillPath?: string;\n    };\n    stderr: {\n      text: string;\n      truncated: boolean;\n      spillPath?: string;\n    };\n    sandbox?: {\n      mode: string;\n      denied: boolean;\n      enforcement?: string;\n      runnerFailed?: boolean;\n    };\n  };\n  cordis_inspect: string;\n  cordis_mount: {\n    id: string;\n    pluginName: string;\n    state: \"pending\" | \"loading\" | \"active\" | \"failed\" | \"disposed\" | \"unloading\";\n    provides: string[];\n    waitingFor: string[];\n  };\n  cordis_unmount: {\n    id: string;\n    pluginName: string;\n  };\n  edit: {\n    path: string;\n    before: string;\n    after: string;\n  };\n  interrupt_agent: {\n    accepted: boolean;\n  };\n  ralph: {\n    runId: string;\n    agentsStarted: number;\n    result: JsonValue;\n  };\n  read: {\n    path: string;\n    offset: number;\n    lines: {\n      number: number;\n      text: string;\n    }[];\n    totalLines: number;\n  };\n  send_message: {\n    messageId: string;\n  };\n  skill: {\n    name: string;\n    provider: string;\n    resourceBase?: {\n      kind: \"directory\";\n      path: string;\n    } | {\n      kind: \"url\";\n      url: string;\n    } | {\n      kind: \"opaque\";\n      description: string;\n    };\n    content: string;\n  };\n  subagent: {\n    kind: \"background\";\n    taskId: string;\n  } | {\n    kind: \"continuable\";\n    subagentId: string;\n  } | {\n    kind: \"foreground\";\n    runId: string;\n    output: JsonValue[];\n  };\n  subagent_fork: {\n    kind: \"background\";\n    taskId: string;\n  } | {\n    kind: \"continuable\";\n    subagentId: string;\n  } | {\n    kind: \"foreground\";\n    runId: string;\n    output: JsonValue[];\n  };\n  task_kill: {\n    outcome: \"cancellation-requested\" | \"already-finished\";\n    task: {\n      id: string;\n      kind: string;\n      label: string;\n      status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n      detail?: string;\n      startedAt: number;\n      finishedAt?: number;\n    };\n  };\n  task_list: ({\n    id: string;\n    kind: string;\n    label: string;\n    status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n    detail?: string;\n    startedAt: number;\n    finishedAt?: number;\n  })[];\n  task_output: {\n    text: string;\n    task: {\n      id: string;\n      kind: string;\n      label: string;\n      status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n      detail?: string;\n      startedAt: number;\n      finishedAt?: number;\n    };\n  };\n  todo_write: {\n    todos: ({\n      content: string;\n      status: \"pending\" | \"in_progress\" | \"completed\";\n    })[];\n    counts: {\n      pending: number;\n      inProgress: number;\n      completed: number;\n    };\n  };\n  workflow: {\n    runId: string;\n    agentsStarted: number;\n    result: JsonValue;\n  };\n  write: {\n    path: string;\n    operation: \"create\" | \"update\";\n    before: string | null;\n    after: string;\n  };\n}\n\ntype ToolName = keyof ToolOutputMap\n\ndeclare class ToolCallError extends Error {\n  readonly name: \"ToolCallError\";\n  readonly toolName: ToolName;\n}\n\ndeclare const tools: {\n  [K in ToolName]: (args: ToolArgsMap[K]) => Promise<ToolOutputMap[K]>;\n}\n```","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under <mode> mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_inspect","description":"Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc.","parameters":{"type":"object","properties":{"what":{"type":"string","description":"Limit the report to one section. Omit for all sections.","enum":["services","plugins","tools","temporary","api","events"]},"name":{"type":"string","description":"Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\"."}}}},{"name":"cordis_mount","description":"Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:<id>]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"JavaScript body returning a temporary Plugin; evaluated now and saved nowhere."}},"required":["code"]}},{"name":"cordis_unmount","description":"Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins.","parameters":{"type":"object","properties":{"id":{"type":"string","description":"The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart."}},"required":["id"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"run_code","description":"Execute a TypeScript program against the available tools. Write the BODY of an async function (erasable syntax only; top-level `await` and `return` work) and call tools as `await tools.name(args)` per the declarations in the system prompt. Only what you print or return comes back — curate it.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"The program: the body of an async TypeScript function."},"description":{"type":"string","description":"Clear, concise description of what this program does in active voice, 5-10 words (shown in the UI). Examples: \"Count TODO markers across packages\"; \"Read failing test and its fixture\"; \"Rename config key in every cordis.yml\"."}},"required":["code","description"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return <value>` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise<any>` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise<any[]>` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise<any[]>` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return <json-value>`)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}}
 {"type":"request/context","seq":9,"time":1785730501507,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}}
 {"type":"assistant/chunk","seq":10,"time":1783957884564,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
 {"type":"assistant/chunk","seq":11,"time":1783957884564,"data":{"turn":1,"step":1,"chunk":{"type":"text-delta","index":0,"text":"DIRECT_CHILD_OK"}}}
 {"type":"assistant/chunk","seq":12,"time":1783957884564,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"DIRECT_CHILD_OK"}}}}
 {"type":"assistant/chunk","seq":13,"time":1785498583897,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}
 {"type":"assistant/chunk","seq":14,"time":1785730501507,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}
-{"type":"assistant/message","seq":15,"time":1785730501507,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"DIRECT_CHILD_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"c5d4e091-9632-4535-af35-097bc74abdd3"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[10,11,12,13,14],"surfaceOp":"append"}
+{"type":"assistant/message","seq":15,"time":1785730501507,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"DIRECT_CHILD_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"391d3467-1155-46f0-b627-456cd0fa32eb"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[10,11,12,13,14],"surfaceOp":"append"}
 {"type":"step/end","seq":16,"time":1785730501507,"data":{"turn":1,"step":1}}
 {"type":"turn/end","seq":17,"time":1785730501507,"data":{"turn":1,"reason":{"kind":"completed"}}}
diff --git a/examples/headless-agent/tests/snapshots/advanced-toolchain/session.2.jsonl b/examples/headless-agent/tests/snapshots/advanced-toolchain/session.2.jsonl
index 8882bda5af..9bc6935f37 100644
--- a/examples/headless-agent/tests/snapshots/advanced-toolchain/session.2.jsonl
+++ b/examples/headless-agent/tests/snapshots/advanced-toolchain/session.2.jsonl
@@ -1,19 +1,19 @@
 {"type":"session","version":0,"id":"33333333-3333-4333-8333-333333333333","createdAt":1783950002000,"cwd":"{{cwd}}","parentSession":"11111111-1111-4111-8111-111111111111","origin":"subagent","delegationDepth":1}
-{"type":"agent/inbox/spliced","seq":0,"time":1785498584048,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Reply with exactly WORKFLOW_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"12fb8b24-9214-4e43-b3d3-47f2af3531f1"}]}}
+{"type":"agent/inbox/spliced","seq":0,"time":1785498584048,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Reply with exactly WORKFLOW_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"471ba696-5559-4115-af5e-092700b5d5ce"}]}}
 {"type":"turn/start","seq":1,"time":1785821454599,"data":{"turn":1}}
 {"type":"agent/inbox/spliced","seq":2,"time":1785821454599,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}
 {"type":"subagent/descriptor","seq":3,"time":1785821454618,"data":{"version":2,"mode":"one-shot","provider":"spawn"}}
 {"type":"step/start","seq":4,"time":1785730501645,"data":{"turn":1,"step":1}}
-{"type":"user/message","seq":5,"time":1785730501645,"data":{"content":[{"type":"text","text":"Reply with exactly WORKFLOW_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"12fb8b24-9214-4e43-b3d3-47f2af3531f1"},"surfaceOp":"append"}
-{"type":"user/message","seq":6,"time":1786358103827,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"14cf8f47-3a7a-4857-a548-02fe407683fb"},"surfaceOp":"append"}
-{"type":"session/title","seq":7,"time":1786358103827,"data":{"title":"Reply with exactly WORKFLOW_CHILD_OK and","messageSeqs":[5],"source":{"kind":"fallback"}}}
-{"type":"request/header","seq":8,"time":1785498584067,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}.\n\nVerify your work by running the code or tests. Keep answers brief and factual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.\n\n## Writing code for run_code\n\nPass `run_code` the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped). Inside the program:\n\n- Call tools as `await tools.name(args)` — quoted access for exotic names: `tools[\"my-tool\"](args)`. Every call resolves to the tool's typed canonical JSON value. Tool arguments must be lossless JSON.\n- A FAILED tool call rejects with `ToolCallError`, whose `toolName` identifies the failed tool and whose `message` is human-readable — `try/catch` it to handle and continue.\n- Independent read-only calls MAY overlap under `Promise.all` (safe calls run concurrently; mutating calls run alone, in submission order). Sequence dependent work with `await`.\n- Emit results with `return` and/or `console.log(...)`. ONLY what you print or return comes back to you — intermediate tool results never enter the conversation, so extract just what you need.\n\nThe available tools:\n\n```ts\ntype JsonValue = null | boolean | number | string | JsonValue[] | { [key: string]: JsonValue }\n\ninterface ToolArgsMap {\n  /** Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under <mode> mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`. */\n  bash: {\n    /** The bash command to execute. */\n    command: string;\n    /** Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\". */\n    description: string;\n    /** Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry. */\n    timeoutMs?: number;\n    /** Working directory for this command. Defaults to the session workspace; a relative path is resolved against it. */\n    workdir?: string;\n    /** Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies. */\n    run_in_background?: boolean;\n  } & Record<string, JsonValue>;\n  /** Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc. */\n  cordis_inspect: {\n    /** Limit the report to one section. Omit for all sections. */\n    what?: \"services\" | \"plugins\" | \"tools\" | \"temporary\" | \"api\" | \"events\";\n    /** Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\". */\n    name?: string;\n  } & Record<string, JsonValue>;\n  /** Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement a normal local, project, or repository Plugin through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:<id>]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime. */\n  cordis_mount: {\n    /** JavaScript body returning a temporary Plugin; evaluated now and saved nowhere. */\n    code: string;\n  } & Record<string, JsonValue>;\n  /** Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins. */\n  cordis_unmount: {\n    /** The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart. */\n    id: string;\n  } & Record<string, JsonValue>;\n  /** Edit an existing UTF-8 text file by replacing literal text. */\n  edit: {\n    /** Path to edit, resolved by the filesystem backend. */\n    file_path: string;\n    /** Literal text to replace. Must match exactly. */\n    old_string: string;\n    /** Literal replacement text. Use an empty string to delete the match. */\n    new_string: string;\n    /** Replace all matches. Defaults to false; when false, old_string must appear exactly once. */\n    replace_all?: boolean;\n  } & Record<string, JsonValue>;\n  /** Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op. */\n  interrupt_agent: {\n    /** The agent id of the running agent to interrupt. */\n    agent_id: string;\n  } & Record<string, JsonValue>;\n  /** Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools. */\n  ralph: {\n    /** The immutable completion objective for every fresh Ralph round. */\n    objective: string;\n    /** Optional positive safe-integer round cap, bounded by the deployment ceiling. */\n    maxRounds?: number;\n  } & Record<string, JsonValue>;\n  /** Read a UTF-8 text file and return line-numbered content. */\n  read: {\n    /** Path to read, resolved by the filesystem backend. */\n    file_path: string;\n    /** 1-based first line to return. Defaults to 1. */\n    offset?: number;\n    /** Maximum number of lines to return. Defaults to 2000. */\n    limit?: number;\n  } & Record<string, JsonValue>;\n  /** Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered. */\n  send_message: {\n    /** The subagent id returned when the background subagent was started. */\n    subagent_id: string;\n    /** The message to deliver to the subagent. */\n    message: string;\n  } & Record<string, JsonValue>;\n  /** Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill. */\n  skill: {\n    /** The exact skill name from the available skills list. */\n    name: string;\n  } & Record<string, JsonValue>;\n  /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */\n  subagent: {\n    /** A short (3-5 word) description of the delegated task, for display. */\n    description: string;\n    /** The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs. */\n    prompt: string;\n    /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */\n    run_in_background?: boolean;\n  } & Record<string, JsonValue>;\n  /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */\n  subagent_fork: {\n    /** A short (3-5 word) description of the delegated task, for display. */\n    description: string;\n    /** The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new. */\n    prompt: string;\n    /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */\n    run_in_background?: boolean;\n  } & Record<string, JsonValue>;\n  /** Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops. */\n  task_kill: {\n    /** Task id returned by the tool that started the background work. */\n    task_id: string;\n    /** Optional short reason, recorded in the log and forwarded to the task. */\n    reason?: string;\n  } & Record<string, JsonValue>;\n  /** List your background tasks (running and finished) with their ids, kinds, and statuses. */\n  task_list: Record<string, JsonValue>;\n  /** Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap. */\n  task_output: {\n    /** Task id returned by the tool that started the background work. */\n    task_id: string;\n    /** Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive. */\n    wait?: boolean;\n    /** Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum. */\n    timeout_ms?: number;\n  } & Record<string, JsonValue>;\n  /** Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished). */\n  todo_write: {\n    /** The COMPLETE task list, replacing any previous list. */\n    todos: ({\n      /** What the task is — a short imperative line. */\n      content: string;\n      /** pending (not started) | in_progress (now) | completed (done). */\n      status: \"pending\" | \"in_progress\" | \"completed\";\n    })[];\n  } & Record<string, JsonValue>;\n  /** Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn. The workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return <value>` — the value must be JSON-serializable and is this tool's result. Script-body hooks: - `agent(prompt, opts?): Promise<any>` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly. - `pipeline(items, ...stages): Promise<any[]>` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages. - `parallel(thunks): Promise<any[]>` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`. - `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim. Misused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`. Constraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes. */\n  workflow: {\n    /** The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return <json-value>`). */\n    script: string;\n    /** The workflow identity block (plain JSON — never code). */\n    meta: {\n      /** Short kebab-case workflow name. */\n      name: string;\n      /** One-line description of what the workflow does. */\n      description: string;\n      /** Optional guidance on when this workflow applies. */\n      whenToUse?: string;\n      /** Optional phase declarations matched by phase() calls. */\n      phases?: ({\n        /** The phase title phase() calls match by exact string. */\n        title: string;\n        /** Optional one-line description of the phase. */\n        detail?: string;\n        /** Optional provider override this phase is expected to use. */\n        provider?: string;\n        /** Optional model override this phase is expected to use. */\n        model?: string;\n      } & Record<string, JsonValue>)[];\n    } & Record<string, JsonValue>;\n    /** Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}). */\n    args?: Record<string, JsonValue>;\n  } & Record<string, JsonValue>;\n  /** Create or fully replace a UTF-8 text file. */\n  write: {\n    /** Path to write, resolved by the filesystem backend. */\n    file_path: string;\n    /** Full UTF-8 text content to write. */\n    content: string;\n  } & Record<string, JsonValue>;\n}\n\ninterface ToolOutputMap {\n  bash: {\n    kind: \"background\";\n    taskId: string;\n  } | {\n    kind: \"foreground\";\n    exitCode: number | null;\n    signal: string | null;\n    timedOut: boolean;\n    aborted: boolean;\n    timeoutMs: number;\n    stdout: {\n      text: string;\n      truncated: boolean;\n      spillPath?: string;\n    };\n    stderr: {\n      text: string;\n      truncated: boolean;\n      spillPath?: string;\n    };\n    sandbox?: {\n      mode: string;\n      denied: boolean;\n      enforcement?: string;\n      runnerFailed?: boolean;\n    };\n  };\n  cordis_inspect: string;\n  cordis_mount: {\n    id: string;\n    pluginName: string;\n    state: \"pending\" | \"loading\" | \"active\" | \"failed\" | \"disposed\" | \"unloading\";\n    provides: string[];\n    waitingFor: string[];\n  };\n  cordis_unmount: {\n    id: string;\n    pluginName: string;\n  };\n  edit: {\n    path: string;\n    before: string;\n    after: string;\n  };\n  interrupt_agent: {\n    accepted: boolean;\n  };\n  ralph: {\n    runId: string;\n    agentsStarted: number;\n    result: JsonValue;\n  };\n  read: {\n    path: string;\n    offset: number;\n    lines: {\n      number: number;\n      text: string;\n    }[];\n    totalLines: number;\n  };\n  send_message: {\n    messageId: string;\n  };\n  skill: {\n    name: string;\n    provider: string;\n    resourceBase?: {\n      kind: \"directory\";\n      path: string;\n    } | {\n      kind: \"url\";\n      url: string;\n    } | {\n      kind: \"opaque\";\n      description: string;\n    };\n    content: string;\n  };\n  subagent: {\n    kind: \"background\";\n    taskId: string;\n  } | {\n    kind: \"continuable\";\n    subagentId: string;\n  } | {\n    kind: \"foreground\";\n    runId: string;\n    output: JsonValue[];\n  };\n  subagent_fork: {\n    kind: \"background\";\n    taskId: string;\n  } | {\n    kind: \"continuable\";\n    subagentId: string;\n  } | {\n    kind: \"foreground\";\n    runId: string;\n    output: JsonValue[];\n  };\n  task_kill: {\n    outcome: \"cancellation-requested\" | \"already-finished\";\n    task: {\n      id: string;\n      kind: string;\n      label: string;\n      status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n      detail?: string;\n      startedAt: number;\n      finishedAt?: number;\n    };\n  };\n  task_list: ({\n    id: string;\n    kind: string;\n    label: string;\n    status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n    detail?: string;\n    startedAt: number;\n    finishedAt?: number;\n  })[];\n  task_output: {\n    text: string;\n    task: {\n      id: string;\n      kind: string;\n      label: string;\n      status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n      detail?: string;\n      startedAt: number;\n      finishedAt?: number;\n    };\n  };\n  todo_write: {\n    todos: ({\n      content: string;\n      status: \"pending\" | \"in_progress\" | \"completed\";\n    })[];\n    counts: {\n      pending: number;\n      inProgress: number;\n      completed: number;\n    };\n  };\n  workflow: {\n    runId: string;\n    agentsStarted: number;\n    result: JsonValue;\n  };\n  write: {\n    path: string;\n    operation: \"create\" | \"update\";\n    before: string | null;\n    after: string;\n  };\n}\n\ntype ToolName = keyof ToolOutputMap\n\ndeclare class ToolCallError extends Error {\n  readonly name: \"ToolCallError\";\n  readonly toolName: ToolName;\n}\n\ndeclare const tools: {\n  [K in ToolName]: (args: ToolArgsMap[K]) => Promise<ToolOutputMap[K]>;\n}\n```","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under <mode> mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_inspect","description":"Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc.","parameters":{"type":"object","properties":{"what":{"type":"string","description":"Limit the report to one section. Omit for all sections.","enum":["services","plugins","tools","temporary","api","events"]},"name":{"type":"string","description":"Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\"."}}}},{"name":"cordis_mount","description":"Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement a normal local, project, or repository Plugin through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:<id>]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"JavaScript body returning a temporary Plugin; evaluated now and saved nowhere."}},"required":["code"]}},{"name":"cordis_unmount","description":"Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins.","parameters":{"type":"object","properties":{"id":{"type":"string","description":"The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart."}},"required":["id"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"run_code","description":"Execute a TypeScript program against the available tools. Write the BODY of an async function (erasable syntax only; top-level `await` and `return` work) and call tools as `await tools.name(args)` per the declarations in the system prompt. Only what you print or return comes back — curate it.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"The program: the body of an async TypeScript function."},"description":{"type":"string","description":"Clear, concise description of what this program does in active voice, 5-10 words (shown in the UI). Examples: \"Count TODO markers across packages\"; \"Read failing test and its fixture\"; \"Rename config key in every cordis.yml\"."}},"required":["code","description"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return <value>` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise<any>` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise<any[]>` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise<any[]>` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return <json-value>`)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}}
+{"type":"user/message","seq":5,"time":1785730501645,"data":{"content":[{"type":"text","text":"Reply with exactly WORKFLOW_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"471ba696-5559-4115-af5e-092700b5d5ce"},"surfaceOp":"append"}
+{"type":"user/message","seq":6,"time":1786373992411,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"e7b80b2f-dbf2-4158-8e26-0b80bd53c5e6"},"surfaceOp":"append"}
+{"type":"session/title","seq":7,"time":1786373992411,"data":{"title":"Reply with exactly WORKFLOW_CHILD_OK and","messageSeqs":[5],"source":{"kind":"fallback"}}}
+{"type":"request/header","seq":8,"time":1785498584067,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}.\n\nVerify your work by running the code or tests. Keep answers brief and factual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.\n\n## Writing code for run_code\n\nPass `run_code` the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped). Inside the program:\n\n- Call tools as `await tools.name(args)` — quoted access for exotic names: `tools[\"my-tool\"](args)`. Every call resolves to the tool's typed canonical JSON value. Tool arguments must be lossless JSON.\n- A FAILED tool call rejects with `ToolCallError`, whose `toolName` identifies the failed tool and whose `message` is human-readable — `try/catch` it to handle and continue.\n- Independent read-only calls MAY overlap under `Promise.all` (safe calls run concurrently; mutating calls run alone, in submission order). Sequence dependent work with `await`.\n- Emit results with `return` and/or `console.log(...)`. ONLY what you print or return comes back to you — intermediate tool results never enter the conversation, so extract just what you need.\n\nThe available tools:\n\n```ts\ntype JsonValue = null | boolean | number | string | JsonValue[] | { [key: string]: JsonValue }\n\ninterface ToolArgsMap {\n  /** Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under <mode> mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`. */\n  bash: {\n    /** The bash command to execute. */\n    command: string;\n    /** Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\". */\n    description: string;\n    /** Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry. */\n    timeoutMs?: number;\n    /** Working directory for this command. Defaults to the session workspace; a relative path is resolved against it. */\n    workdir?: string;\n    /** Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies. */\n    run_in_background?: boolean;\n  } & Record<string, JsonValue>;\n  /** Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc. */\n  cordis_inspect: {\n    /** Limit the report to one section. Omit for all sections. */\n    what?: \"services\" | \"plugins\" | \"tools\" | \"temporary\" | \"api\" | \"events\";\n    /** Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\". */\n    name?: string;\n  } & Record<string, JsonValue>;\n  /** Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:<id>]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime. */\n  cordis_mount: {\n    /** JavaScript body returning a temporary Plugin; evaluated now and saved nowhere. */\n    code: string;\n  } & Record<string, JsonValue>;\n  /** Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins. */\n  cordis_unmount: {\n    /** The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart. */\n    id: string;\n  } & Record<string, JsonValue>;\n  /** Edit an existing UTF-8 text file by replacing literal text. */\n  edit: {\n    /** Path to edit, resolved by the filesystem backend. */\n    file_path: string;\n    /** Literal text to replace. Must match exactly. */\n    old_string: string;\n    /** Literal replacement text. Use an empty string to delete the match. */\n    new_string: string;\n    /** Replace all matches. Defaults to false; when false, old_string must appear exactly once. */\n    replace_all?: boolean;\n  } & Record<string, JsonValue>;\n  /** Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op. */\n  interrupt_agent: {\n    /** The agent id of the running agent to interrupt. */\n    agent_id: string;\n  } & Record<string, JsonValue>;\n  /** Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools. */\n  ralph: {\n    /** The immutable completion objective for every fresh Ralph round. */\n    objective: string;\n    /** Optional positive safe-integer round cap, bounded by the deployment ceiling. */\n    maxRounds?: number;\n  } & Record<string, JsonValue>;\n  /** Read a UTF-8 text file and return line-numbered content. */\n  read: {\n    /** Path to read, resolved by the filesystem backend. */\n    file_path: string;\n    /** 1-based first line to return. Defaults to 1. */\n    offset?: number;\n    /** Maximum number of lines to return. Defaults to 2000. */\n    limit?: number;\n  } & Record<string, JsonValue>;\n  /** Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered. */\n  send_message: {\n    /** The subagent id returned when the background subagent was started. */\n    subagent_id: string;\n    /** The message to deliver to the subagent. */\n    message: string;\n  } & Record<string, JsonValue>;\n  /** Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill. */\n  skill: {\n    /** The exact skill name from the available skills list. */\n    name: string;\n  } & Record<string, JsonValue>;\n  /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work. */\n  subagent: {\n    /** A short (3-5 word) description of the delegated task, for display. */\n    description: string;\n    /** The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs. */\n    prompt: string;\n    /** Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message. */\n    run_in_background?: boolean;\n  } & Record<string, JsonValue>;\n  /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. */\n  subagent_fork: {\n    /** A short (3-5 word) description of the delegated task, for display. */\n    description: string;\n    /** The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new. */\n    prompt: string;\n  } & Record<string, JsonValue>;\n  /** Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops. */\n  task_kill: {\n    /** Task id returned by the tool that started the background work. */\n    task_id: string;\n    /** Optional short reason, recorded in the log and forwarded to the task. */\n    reason?: string;\n  } & Record<string, JsonValue>;\n  /** List your background tasks (running and finished) with their ids, kinds, and statuses. */\n  task_list: Record<string, JsonValue>;\n  /** Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap. */\n  task_output: {\n    /** Task id returned by the tool that started the background work. */\n    task_id: string;\n    /** Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive. */\n    wait?: boolean;\n    /** Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum. */\n    timeout_ms?: number;\n  } & Record<string, JsonValue>;\n  /** Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished). */\n  todo_write: {\n    /** The COMPLETE task list, replacing any previous list. */\n    todos: ({\n      /** What the task is — a short imperative line. */\n      content: string;\n      /** pending (not started) | in_progress (now) | completed (done). */\n      status: \"pending\" | \"in_progress\" | \"completed\";\n    })[];\n  } & Record<string, JsonValue>;\n  /** Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn. The workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return <value>` — the value must be JSON-serializable and is this tool's result. Script-body hooks: - `agent(prompt, opts?): Promise<any>` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly. - `pipeline(items, ...stages): Promise<any[]>` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages. - `parallel(thunks): Promise<any[]>` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`. - `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim. Misused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`. Constraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes. */\n  workflow: {\n    /** The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return <json-value>`). */\n    script: string;\n    /** The workflow identity block (plain JSON — never code). */\n    meta: {\n      /** Short kebab-case workflow name. */\n      name: string;\n      /** One-line description of what the workflow does. */\n      description: string;\n      /** Optional guidance on when this workflow applies. */\n      whenToUse?: string;\n      /** Optional phase declarations matched by phase() calls. */\n      phases?: ({\n        /** The phase title phase() calls match by exact string. */\n        title: string;\n        /** Optional one-line description of the phase. */\n        detail?: string;\n        /** Optional provider override this phase is expected to use. */\n        provider?: string;\n        /** Optional model override this phase is expected to use. */\n        model?: string;\n      } & Record<string, JsonValue>)[];\n    } & Record<string, JsonValue>;\n    /** Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}). */\n    args?: Record<string, JsonValue>;\n  } & Record<string, JsonValue>;\n  /** Create or fully replace a UTF-8 text file. */\n  write: {\n    /** Path to write, resolved by the filesystem backend. */\n    file_path: string;\n    /** Full UTF-8 text content to write. */\n    content: string;\n  } & Record<string, JsonValue>;\n}\n\ninterface ToolOutputMap {\n  bash: {\n    kind: \"background\";\n    taskId: string;\n  } | {\n    kind: \"foreground\";\n    exitCode: number | null;\n    signal: string | null;\n    timedOut: boolean;\n    aborted: boolean;\n    timeoutMs: number;\n    stdout: {\n      text: string;\n      truncated: boolean;\n      spillPath?: string;\n    };\n    stderr: {\n      text: string;\n      truncated: boolean;\n      spillPath?: string;\n    };\n    sandbox?: {\n      mode: string;\n      denied: boolean;\n      enforcement?: string;\n      runnerFailed?: boolean;\n    };\n  };\n  cordis_inspect: string;\n  cordis_mount: {\n    id: string;\n    pluginName: string;\n    state: \"pending\" | \"loading\" | \"active\" | \"failed\" | \"disposed\" | \"unloading\";\n    provides: string[];\n    waitingFor: string[];\n  };\n  cordis_unmount: {\n    id: string;\n    pluginName: string;\n  };\n  edit: {\n    path: string;\n    before: string;\n    after: string;\n  };\n  interrupt_agent: {\n    accepted: boolean;\n  };\n  ralph: {\n    runId: string;\n    agentsStarted: number;\n    result: JsonValue;\n  };\n  read: {\n    path: string;\n    offset: number;\n    lines: {\n      number: number;\n      text: string;\n    }[];\n    totalLines: number;\n  };\n  send_message: {\n    messageId: string;\n  };\n  skill: {\n    name: string;\n    provider: string;\n    resourceBase?: {\n      kind: \"directory\";\n      path: string;\n    } | {\n      kind: \"url\";\n      url: string;\n    } | {\n      kind: \"opaque\";\n      description: string;\n    };\n    content: string;\n  };\n  subagent: {\n    kind: \"background\";\n    taskId: string;\n  } | {\n    kind: \"continuable\";\n    subagentId: string;\n  } | {\n    kind: \"foreground\";\n    runId: string;\n    output: JsonValue[];\n  };\n  subagent_fork: {\n    kind: \"background\";\n    taskId: string;\n  } | {\n    kind: \"continuable\";\n    subagentId: string;\n  } | {\n    kind: \"foreground\";\n    runId: string;\n    output: JsonValue[];\n  };\n  task_kill: {\n    outcome: \"cancellation-requested\" | \"already-finished\";\n    task: {\n      id: string;\n      kind: string;\n      label: string;\n      status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n      detail?: string;\n      startedAt: number;\n      finishedAt?: number;\n    };\n  };\n  task_list: ({\n    id: string;\n    kind: string;\n    label: string;\n    status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n    detail?: string;\n    startedAt: number;\n    finishedAt?: number;\n  })[];\n  task_output: {\n    text: string;\n    task: {\n      id: string;\n      kind: string;\n      label: string;\n      status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n      detail?: string;\n      startedAt: number;\n      finishedAt?: number;\n    };\n  };\n  todo_write: {\n    todos: ({\n      content: string;\n      status: \"pending\" | \"in_progress\" | \"completed\";\n    })[];\n    counts: {\n      pending: number;\n      inProgress: number;\n      completed: number;\n    };\n  };\n  workflow: {\n    runId: string;\n    agentsStarted: number;\n    result: JsonValue;\n  };\n  write: {\n    path: string;\n    operation: \"create\" | \"update\";\n    before: string | null;\n    after: string;\n  };\n}\n\ntype ToolName = keyof ToolOutputMap\n\ndeclare class ToolCallError extends Error {\n  readonly name: \"ToolCallError\";\n  readonly toolName: ToolName;\n}\n\ndeclare const tools: {\n  [K in ToolName]: (args: ToolArgsMap[K]) => Promise<ToolOutputMap[K]>;\n}\n```","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under <mode> mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_inspect","description":"Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc.","parameters":{"type":"object","properties":{"what":{"type":"string","description":"Limit the report to one section. Omit for all sections.","enum":["services","plugins","tools","temporary","api","events"]},"name":{"type":"string","description":"Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\"."}}}},{"name":"cordis_mount","description":"Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:<id>]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"JavaScript body returning a temporary Plugin; evaluated now and saved nowhere."}},"required":["code"]}},{"name":"cordis_unmount","description":"Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins.","parameters":{"type":"object","properties":{"id":{"type":"string","description":"The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart."}},"required":["id"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"run_code","description":"Execute a TypeScript program against the available tools. Write the BODY of an async function (erasable syntax only; top-level `await` and `return` work) and call tools as `await tools.name(args)` per the declarations in the system prompt. Only what you print or return comes back — curate it.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"The program: the body of an async TypeScript function."},"description":{"type":"string","description":"Clear, concise description of what this program does in active voice, 5-10 words (shown in the UI). Examples: \"Count TODO markers across packages\"; \"Read failing test and its fixture\"; \"Rename config key in every cordis.yml\"."}},"required":["code","description"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return <value>` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise<any>` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise<any[]>` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise<any[]>` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return <json-value>`)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}}
 {"type":"request/context","seq":9,"time":1785730501646,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}}
 {"type":"assistant/chunk","seq":10,"time":1783957884701,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
 {"type":"assistant/chunk","seq":11,"time":1783957884701,"data":{"turn":1,"step":1,"chunk":{"type":"text-delta","index":0,"text":"WORKFLOW_CHILD_OK"}}}
 {"type":"assistant/chunk","seq":12,"time":1783957884701,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"WORKFLOW_CHILD_OK"}}}}
 {"type":"assistant/chunk","seq":13,"time":1785498584067,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}
 {"type":"assistant/chunk","seq":14,"time":1785730501646,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}
-{"type":"assistant/message","seq":15,"time":1785730501646,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"WORKFLOW_CHILD_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"bb9c39a9-3239-4ee1-939a-bab0046e3028"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[10,11,12,13,14],"surfaceOp":"append"}
+{"type":"assistant/message","seq":15,"time":1785730501646,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"WORKFLOW_CHILD_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"fd781f8d-2c38-4b32-846b-74b2726f3e78"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[10,11,12,13,14],"surfaceOp":"append"}
 {"type":"step/end","seq":16,"time":1785730501646,"data":{"turn":1,"step":1}}
 {"type":"turn/end","seq":17,"time":1785730501646,"data":{"turn":1,"reason":{"kind":"completed"}}}
diff --git a/examples/headless-agent/tests/snapshots/advanced-toolchain/session.jsonl b/examples/headless-agent/tests/snapshots/advanced-toolchain/session.jsonl
index 8f0a211969..7b14e5447f 100644
--- a/examples/headless-agent/tests/snapshots/advanced-toolchain/session.jsonl
+++ b/examples/headless-agent/tests/snapshots/advanced-toolchain/session.jsonl
@@ -1,20 +1,20 @@
 {"type":"session","version":0,"id":"11111111-1111-4111-8111-111111111111","createdAt":1783950000000,"cwd":"{{cwd}}","delegationDepth":0}
-{"type":"agent/inbox/spliced","seq":0,"time":1785498583746,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Run this advanced flow exactly once: try a no-op temporary Cordis Plugin named snapshot-marker; use run_code to inspect the live temporary Plugins through tools.cordis_inspect; delegate once to a direct spawn child; run one workflow that delegates to another spawn child; stop dyn-1; then reply with exactly ADVANCED_HEADLESS_OK."}],"source":{"kind":"user"},"role":"user","id":"2b5f9025-324a-4c76-b883-4af3b5c3060a"}]}}
+{"type":"agent/inbox/spliced","seq":0,"time":1785498583746,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Run this advanced flow exactly once: try a no-op temporary Cordis Plugin named snapshot-marker; use run_code to inspect the live temporary Plugins through tools.cordis_inspect; delegate once to a direct spawn child; run one workflow that delegates to another spawn child; stop dyn-1; then reply with exactly ADVANCED_HEADLESS_OK."}],"source":{"kind":"user"},"role":"user","id":"fa8e6017-366e-4c5b-a8ff-0c84471cffe8"}]}}
 {"type":"turn/start","seq":1,"time":1785821454304,"data":{"turn":1}}
 {"type":"agent/inbox/spliced","seq":2,"time":1785821454304,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}
 {"type":"step/start","seq":3,"time":1783957884486,"data":{"turn":1,"step":1}}
-{"type":"user/message","seq":4,"time":1785498583779,"data":{"content":[{"type":"text","text":"Run this advanced flow exactly once: try a no-op temporary Cordis Plugin named snapshot-marker; use run_code to inspect the live temporary Plugins through tools.cordis_inspect; delegate once to a direct spawn child; run one workflow that delegates to another spawn child; stop dyn-1; then reply with exactly ADVANCED_HEADLESS_OK."}],"source":{"kind":"user"},"role":"user","id":"2b5f9025-324a-4c76-b883-4af3b5c3060a"},"surfaceOp":"append"}
+{"type":"user/message","seq":4,"time":1785498583779,"data":{"content":[{"type":"text","text":"Run this advanced flow exactly once: try a no-op temporary Cordis Plugin named snapshot-marker; use run_code to inspect the live temporary Plugins through tools.cordis_inspect; delegate once to a direct spawn child; run one workflow that delegates to another spawn child; stop dyn-1; then reply with exactly ADVANCED_HEADLESS_OK."}],"source":{"kind":"user"},"role":"user","id":"fa8e6017-366e-4c5b-a8ff-0c84471cffe8"},"surfaceOp":"append"}
 {"type":"session/title","seq":5,"time":1785498583779,"data":{"title":"Run this advanced flow exactly","messageSeqs":[4],"source":{"kind":"fallback"}}}
-{"type":"request/header","seq":6,"time":1785498583782,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}.\n\nVerify your work by running the code or tests. Keep answers brief and factual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.\n\n## Writing code for run_code\n\nPass `run_code` the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped). Inside the program:\n\n- Call tools as `await tools.name(args)` — quoted access for exotic names: `tools[\"my-tool\"](args)`. Every call resolves to the tool's typed canonical JSON value. Tool arguments must be lossless JSON.\n- A FAILED tool call rejects with `ToolCallError`, whose `toolName` identifies the failed tool and whose `message` is human-readable — `try/catch` it to handle and continue.\n- Independent read-only calls MAY overlap under `Promise.all` (safe calls run concurrently; mutating calls run alone, in submission order). Sequence dependent work with `await`.\n- Emit results with `return` and/or `console.log(...)`. ONLY what you print or return comes back to you — intermediate tool results never enter the conversation, so extract just what you need.\n\nThe available tools:\n\n```ts\ntype JsonValue = null | boolean | number | string | JsonValue[] | { [key: string]: JsonValue }\n\ninterface ToolArgsMap {\n  /** Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under <mode> mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`. */\n  bash: {\n    /** The bash command to execute. */\n    command: string;\n    /** Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\". */\n    description: string;\n    /** Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry. */\n    timeoutMs?: number;\n    /** Working directory for this command. Defaults to the session workspace; a relative path is resolved against it. */\n    workdir?: string;\n    /** Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies. */\n    run_in_background?: boolean;\n  } & Record<string, JsonValue>;\n  /** Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc. */\n  cordis_inspect: {\n    /** Limit the report to one section. Omit for all sections. */\n    what?: \"services\" | \"plugins\" | \"tools\" | \"temporary\" | \"api\" | \"events\";\n    /** Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\". */\n    name?: string;\n  } & Record<string, JsonValue>;\n  /** Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement a normal local, project, or repository Plugin through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:<id>]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime. */\n  cordis_mount: {\n    /** JavaScript body returning a temporary Plugin; evaluated now and saved nowhere. */\n    code: string;\n  } & Record<string, JsonValue>;\n  /** Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins. */\n  cordis_unmount: {\n    /** The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart. */\n    id: string;\n  } & Record<string, JsonValue>;\n  /** Edit an existing UTF-8 text file by replacing literal text. */\n  edit: {\n    /** Path to edit, resolved by the filesystem backend. */\n    file_path: string;\n    /** Literal text to replace. Must match exactly. */\n    old_string: string;\n    /** Literal replacement text. Use an empty string to delete the match. */\n    new_string: string;\n    /** Replace all matches. Defaults to false; when false, old_string must appear exactly once. */\n    replace_all?: boolean;\n  } & Record<string, JsonValue>;\n  /** Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op. */\n  interrupt_agent: {\n    /** The agent id of the running agent to interrupt. */\n    agent_id: string;\n  } & Record<string, JsonValue>;\n  /** Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools. */\n  ralph: {\n    /** The immutable completion objective for every fresh Ralph round. */\n    objective: string;\n    /** Optional positive safe-integer round cap, bounded by the deployment ceiling. */\n    maxRounds?: number;\n  } & Record<string, JsonValue>;\n  /** Read a UTF-8 text file and return line-numbered content. */\n  read: {\n    /** Path to read, resolved by the filesystem backend. */\n    file_path: string;\n    /** 1-based first line to return. Defaults to 1. */\n    offset?: number;\n    /** Maximum number of lines to return. Defaults to 2000. */\n    limit?: number;\n  } & Record<string, JsonValue>;\n  /** Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered. */\n  send_message: {\n    /** The subagent id returned when the background subagent was started. */\n    subagent_id: string;\n    /** The message to deliver to the subagent. */\n    message: string;\n  } & Record<string, JsonValue>;\n  /** Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill. */\n  skill: {\n    /** The exact skill name from the available skills list. */\n    name: string;\n  } & Record<string, JsonValue>;\n  /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */\n  subagent: {\n    /** A short (3-5 word) description of the delegated task, for display. */\n    description: string;\n    /** The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs. */\n    prompt: string;\n    /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */\n    run_in_background?: boolean;\n  } & Record<string, JsonValue>;\n  /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */\n  subagent_fork: {\n    /** A short (3-5 word) description of the delegated task, for display. */\n    description: string;\n    /** The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new. */\n    prompt: string;\n    /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */\n    run_in_background?: boolean;\n  } & Record<string, JsonValue>;\n  /** Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops. */\n  task_kill: {\n    /** Task id returned by the tool that started the background work. */\n    task_id: string;\n    /** Optional short reason, recorded in the log and forwarded to the task. */\n    reason?: string;\n  } & Record<string, JsonValue>;\n  /** List your background tasks (running and finished) with their ids, kinds, and statuses. */\n  task_list: Record<string, JsonValue>;\n  /** Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap. */\n  task_output: {\n    /** Task id returned by the tool that started the background work. */\n    task_id: string;\n    /** Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive. */\n    wait?: boolean;\n    /** Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum. */\n    timeout_ms?: number;\n  } & Record<string, JsonValue>;\n  /** Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished). */\n  todo_write: {\n    /** The COMPLETE task list, replacing any previous list. */\n    todos: ({\n      /** What the task is — a short imperative line. */\n      content: string;\n      /** pending (not started) | in_progress (now) | completed (done). */\n      status: \"pending\" | \"in_progress\" | \"completed\";\n    })[];\n  } & Record<string, JsonValue>;\n  /** Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn. The workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return <value>` — the value must be JSON-serializable and is this tool's result. Script-body hooks: - `agent(prompt, opts?): Promise<any>` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly. - `pipeline(items, ...stages): Promise<any[]>` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages. - `parallel(thunks): Promise<any[]>` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`. - `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim. Misused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`. Constraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes. */\n  workflow: {\n    /** The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return <json-value>`). */\n    script: string;\n    /** The workflow identity block (plain JSON — never code). */\n    meta: {\n      /** Short kebab-case workflow name. */\n      name: string;\n      /** One-line description of what the workflow does. */\n      description: string;\n      /** Optional guidance on when this workflow applies. */\n      whenToUse?: string;\n      /** Optional phase declarations matched by phase() calls. */\n      phases?: ({\n        /** The phase title phase() calls match by exact string. */\n        title: string;\n        /** Optional one-line description of the phase. */\n        detail?: string;\n        /** Optional provider override this phase is expected to use. */\n        provider?: string;\n        /** Optional model override this phase is expected to use. */\n        model?: string;\n      } & Record<string, JsonValue>)[];\n    } & Record<string, JsonValue>;\n    /** Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}). */\n    args?: Record<string, JsonValue>;\n  } & Record<string, JsonValue>;\n  /** Create or fully replace a UTF-8 text file. */\n  write: {\n    /** Path to write, resolved by the filesystem backend. */\n    file_path: string;\n    /** Full UTF-8 text content to write. */\n    content: string;\n  } & Record<string, JsonValue>;\n}\n\ninterface ToolOutputMap {\n  bash: {\n    kind: \"background\";\n    taskId: string;\n  } | {\n    kind: \"foreground\";\n    exitCode: number | null;\n    signal: string | null;\n    timedOut: boolean;\n    aborted: boolean;\n    timeoutMs: number;\n    stdout: {\n      text: string;\n      truncated: boolean;\n      spillPath?: string;\n    };\n    stderr: {\n      text: string;\n      truncated: boolean;\n      spillPath?: string;\n    };\n    sandbox?: {\n      mode: string;\n      denied: boolean;\n      enforcement?: string;\n      runnerFailed?: boolean;\n    };\n  };\n  cordis_inspect: string;\n  cordis_mount: {\n    id: string;\n    pluginName: string;\n    state: \"pending\" | \"loading\" | \"active\" | \"failed\" | \"disposed\" | \"unloading\";\n    provides: string[];\n    waitingFor: string[];\n  };\n  cordis_unmount: {\n    id: string;\n    pluginName: string;\n  };\n  edit: {\n    path: string;\n    before: string;\n    after: string;\n  };\n  interrupt_agent: {\n    accepted: boolean;\n  };\n  ralph: {\n    runId: string;\n    agentsStarted: number;\n    result: JsonValue;\n  };\n  read: {\n    path: string;\n    offset: number;\n    lines: {\n      number: number;\n      text: string;\n    }[];\n    totalLines: number;\n  };\n  send_message: {\n    messageId: string;\n  };\n  skill: {\n    name: string;\n    provider: string;\n    resourceBase?: {\n      kind: \"directory\";\n      path: string;\n    } | {\n      kind: \"url\";\n      url: string;\n    } | {\n      kind: \"opaque\";\n      description: string;\n    };\n    content: string;\n  };\n  subagent: {\n    kind: \"background\";\n    taskId: string;\n  } | {\n    kind: \"continuable\";\n    subagentId: string;\n  } | {\n    kind: \"foreground\";\n    runId: string;\n    output: JsonValue[];\n  };\n  subagent_fork: {\n    kind: \"background\";\n    taskId: string;\n  } | {\n    kind: \"continuable\";\n    subagentId: string;\n  } | {\n    kind: \"foreground\";\n    runId: string;\n    output: JsonValue[];\n  };\n  task_kill: {\n    outcome: \"cancellation-requested\" | \"already-finished\";\n    task: {\n      id: string;\n      kind: string;\n      label: string;\n      status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n      detail?: string;\n      startedAt: number;\n      finishedAt?: number;\n    };\n  };\n  task_list: ({\n    id: string;\n    kind: string;\n    label: string;\n    status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n    detail?: string;\n    startedAt: number;\n    finishedAt?: number;\n  })[];\n  task_output: {\n    text: string;\n    task: {\n      id: string;\n      kind: string;\n      label: string;\n      status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n      detail?: string;\n      startedAt: number;\n      finishedAt?: number;\n    };\n  };\n  todo_write: {\n    todos: ({\n      content: string;\n      status: \"pending\" | \"in_progress\" | \"completed\";\n    })[];\n    counts: {\n      pending: number;\n      inProgress: number;\n      completed: number;\n    };\n  };\n  workflow: {\n    runId: string;\n    agentsStarted: number;\n    result: JsonValue;\n  };\n  write: {\n    path: string;\n    operation: \"create\" | \"update\";\n    before: string | null;\n    after: string;\n  };\n}\n\ntype ToolName = keyof ToolOutputMap\n\ndeclare class ToolCallError extends Error {\n  readonly name: \"ToolCallError\";\n  readonly toolName: ToolName;\n}\n\ndeclare const tools: {\n  [K in ToolName]: (args: ToolArgsMap[K]) => Promise<ToolOutputMap[K]>;\n}\n```","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under <mode> mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_inspect","description":"Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc.","parameters":{"type":"object","properties":{"what":{"type":"string","description":"Limit the report to one section. Omit for all sections.","enum":["services","plugins","tools","temporary","api","events"]},"name":{"type":"string","description":"Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\"."}}}},{"name":"cordis_mount","description":"Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement a normal local, project, or repository Plugin through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:<id>]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"JavaScript body returning a temporary Plugin; evaluated now and saved nowhere."}},"required":["code"]}},{"name":"cordis_unmount","description":"Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins.","parameters":{"type":"object","properties":{"id":{"type":"string","description":"The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart."}},"required":["id"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"run_code","description":"Execute a TypeScript program against the available tools. Write the BODY of an async function (erasable syntax only; top-level `await` and `return` work) and call tools as `await tools.name(args)` per the declarations in the system prompt. Only what you print or return comes back — curate it.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"The program: the body of an async TypeScript function."},"description":{"type":"string","description":"Clear, concise description of what this program does in active voice, 5-10 words (shown in the UI). Examples: \"Count TODO markers across packages\"; \"Read failing test and its fixture\"; \"Rename config key in every cordis.yml\"."}},"required":["code","description"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return <value>` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise<any>` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise<any[]>` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise<any[]>` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return <json-value>`)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}}
+{"type":"request/header","seq":6,"time":1785498583782,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}.\n\nVerify your work by running the code or tests. Keep answers brief and factual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.\n\n## Writing code for run_code\n\nPass `run_code` the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped). Inside the program:\n\n- Call tools as `await tools.name(args)` — quoted access for exotic names: `tools[\"my-tool\"](args)`. Every call resolves to the tool's typed canonical JSON value. Tool arguments must be lossless JSON.\n- A FAILED tool call rejects with `ToolCallError`, whose `toolName` identifies the failed tool and whose `message` is human-readable — `try/catch` it to handle and continue.\n- Independent read-only calls MAY overlap under `Promise.all` (safe calls run concurrently; mutating calls run alone, in submission order). Sequence dependent work with `await`.\n- Emit results with `return` and/or `console.log(...)`. ONLY what you print or return comes back to you — intermediate tool results never enter the conversation, so extract just what you need.\n\nThe available tools:\n\n```ts\ntype JsonValue = null | boolean | number | string | JsonValue[] | { [key: string]: JsonValue }\n\ninterface ToolArgsMap {\n  /** Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under <mode> mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`. */\n  bash: {\n    /** The bash command to execute. */\n    command: string;\n    /** Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\". */\n    description: string;\n    /** Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry. */\n    timeoutMs?: number;\n    /** Working directory for this command. Defaults to the session workspace; a relative path is resolved against it. */\n    workdir?: string;\n    /** Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies. */\n    run_in_background?: boolean;\n  } & Record<string, JsonValue>;\n  /** Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc. */\n  cordis_inspect: {\n    /** Limit the report to one section. Omit for all sections. */\n    what?: \"services\" | \"plugins\" | \"tools\" | \"temporary\" | \"api\" | \"events\";\n    /** Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\". */\n    name?: string;\n  } & Record<string, JsonValue>;\n  /** Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:<id>]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime. */\n  cordis_mount: {\n    /** JavaScript body returning a temporary Plugin; evaluated now and saved nowhere. */\n    code: string;\n  } & Record<string, JsonValue>;\n  /** Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins. */\n  cordis_unmount: {\n    /** The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart. */\n    id: string;\n  } & Record<string, JsonValue>;\n  /** Edit an existing UTF-8 text file by replacing literal text. */\n  edit: {\n    /** Path to edit, resolved by the filesystem backend. */\n    file_path: string;\n    /** Literal text to replace. Must match exactly. */\n    old_string: string;\n    /** Literal replacement text. Use an empty string to delete the match. */\n    new_string: string;\n    /** Replace all matches. Defaults to false; when false, old_string must appear exactly once. */\n    replace_all?: boolean;\n  } & Record<string, JsonValue>;\n  /** Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op. */\n  interrupt_agent: {\n    /** The agent id of the running agent to interrupt. */\n    agent_id: string;\n  } & Record<string, JsonValue>;\n  /** Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools. */\n  ralph: {\n    /** The immutable completion objective for every fresh Ralph round. */\n    objective: string;\n    /** Optional positive safe-integer round cap, bounded by the deployment ceiling. */\n    maxRounds?: number;\n  } & Record<string, JsonValue>;\n  /** Read a UTF-8 text file and return line-numbered content. */\n  read: {\n    /** Path to read, resolved by the filesystem backend. */\n    file_path: string;\n    /** 1-based first line to return. Defaults to 1. */\n    offset?: number;\n    /** Maximum number of lines to return. Defaults to 2000. */\n    limit?: number;\n  } & Record<string, JsonValue>;\n  /** Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered. */\n  send_message: {\n    /** The subagent id returned when the background subagent was started. */\n    subagent_id: string;\n    /** The message to deliver to the subagent. */\n    message: string;\n  } & Record<string, JsonValue>;\n  /** Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill. */\n  skill: {\n    /** The exact skill name from the available skills list. */\n    name: string;\n  } & Record<string, JsonValue>;\n  /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work. */\n  subagent: {\n    /** A short (3-5 word) description of the delegated task, for display. */\n    description: string;\n    /** The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs. */\n    prompt: string;\n    /** Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message. */\n    run_in_background?: boolean;\n  } & Record<string, JsonValue>;\n  /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. */\n  subagent_fork: {\n    /** A short (3-5 word) description of the delegated task, for display. */\n    description: string;\n    /** The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new. */\n    prompt: string;\n  } & Record<string, JsonValue>;\n  /** Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops. */\n  task_kill: {\n    /** Task id returned by the tool that started the background work. */\n    task_id: string;\n    /** Optional short reason, recorded in the log and forwarded to the task. */\n    reason?: string;\n  } & Record<string, JsonValue>;\n  /** List your background tasks (running and finished) with their ids, kinds, and statuses. */\n  task_list: Record<string, JsonValue>;\n  /** Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap. */\n  task_output: {\n    /** Task id returned by the tool that started the background work. */\n    task_id: string;\n    /** Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive. */\n    wait?: boolean;\n    /** Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum. */\n    timeout_ms?: number;\n  } & Record<string, JsonValue>;\n  /** Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished). */\n  todo_write: {\n    /** The COMPLETE task list, replacing any previous list. */\n    todos: ({\n      /** What the task is — a short imperative line. */\n      content: string;\n      /** pending (not started) | in_progress (now) | completed (done). */\n      status: \"pending\" | \"in_progress\" | \"completed\";\n    })[];\n  } & Record<string, JsonValue>;\n  /** Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn. The workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return <value>` — the value must be JSON-serializable and is this tool's result. Script-body hooks: - `agent(prompt, opts?): Promise<any>` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly. - `pipeline(items, ...stages): Promise<any[]>` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages. - `parallel(thunks): Promise<any[]>` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`. - `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim. Misused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`. Constraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes. */\n  workflow: {\n    /** The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return <json-value>`). */\n    script: string;\n    /** The workflow identity block (plain JSON — never code). */\n    meta: {\n      /** Short kebab-case workflow name. */\n      name: string;\n      /** One-line description of what the workflow does. */\n      description: string;\n      /** Optional guidance on when this workflow applies. */\n      whenToUse?: string;\n      /** Optional phase declarations matched by phase() calls. */\n      phases?: ({\n        /** The phase title phase() calls match by exact string. */\n        title: string;\n        /** Optional one-line description of the phase. */\n        detail?: string;\n        /** Optional provider override this phase is expected to use. */\n        provider?: string;\n        /** Optional model override this phase is expected to use. */\n        model?: string;\n      } & Record<string, JsonValue>)[];\n    } & Record<string, JsonValue>;\n    /** Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}). */\n    args?: Record<string, JsonValue>;\n  } & Record<string, JsonValue>;\n  /** Create or fully replace a UTF-8 text file. */\n  write: {\n    /** Path to write, resolved by the filesystem backend. */\n    file_path: string;\n    /** Full UTF-8 text content to write. */\n    content: string;\n  } & Record<string, JsonValue>;\n}\n\ninterface ToolOutputMap {\n  bash: {\n    kind: \"background\";\n    taskId: string;\n  } | {\n    kind: \"foreground\";\n    exitCode: number | null;\n    signal: string | null;\n    timedOut: boolean;\n    aborted: boolean;\n    timeoutMs: number;\n    stdout: {\n      text: string;\n      truncated: boolean;\n      spillPath?: string;\n    };\n    stderr: {\n      text: string;\n      truncated: boolean;\n      spillPath?: string;\n    };\n    sandbox?: {\n      mode: string;\n      denied: boolean;\n      enforcement?: string;\n      runnerFailed?: boolean;\n    };\n  };\n  cordis_inspect: string;\n  cordis_mount: {\n    id: string;\n    pluginName: string;\n    state: \"pending\" | \"loading\" | \"active\" | \"failed\" | \"disposed\" | \"unloading\";\n    provides: string[];\n    waitingFor: string[];\n  };\n  cordis_unmount: {\n    id: string;\n    pluginName: string;\n  };\n  edit: {\n    path: string;\n    before: string;\n    after: string;\n  };\n  interrupt_agent: {\n    accepted: boolean;\n  };\n  ralph: {\n    runId: string;\n    agentsStarted: number;\n    result: JsonValue;\n  };\n  read: {\n    path: string;\n    offset: number;\n    lines: {\n      number: number;\n      text: string;\n    }[];\n    totalLines: number;\n  };\n  send_message: {\n    messageId: string;\n  };\n  skill: {\n    name: string;\n    provider: string;\n    resourceBase?: {\n      kind: \"directory\";\n      path: string;\n    } | {\n      kind: \"url\";\n      url: string;\n    } | {\n      kind: \"opaque\";\n      description: string;\n    };\n    content: string;\n  };\n  subagent: {\n    kind: \"background\";\n    taskId: string;\n  } | {\n    kind: \"continuable\";\n    subagentId: string;\n  } | {\n    kind: \"foreground\";\n    runId: string;\n    output: JsonValue[];\n  };\n  subagent_fork: {\n    kind: \"background\";\n    taskId: string;\n  } | {\n    kind: \"continuable\";\n    subagentId: string;\n  } | {\n    kind: \"foreground\";\n    runId: string;\n    output: JsonValue[];\n  };\n  task_kill: {\n    outcome: \"cancellation-requested\" | \"already-finished\";\n    task: {\n      id: string;\n      kind: string;\n      label: string;\n      status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n      detail?: string;\n      startedAt: number;\n      finishedAt?: number;\n    };\n  };\n  task_list: ({\n    id: string;\n    kind: string;\n    label: string;\n    status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n    detail?: string;\n    startedAt: number;\n    finishedAt?: number;\n  })[];\n  task_output: {\n    text: string;\n    task: {\n      id: string;\n      kind: string;\n      label: string;\n      status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n      detail?: string;\n      startedAt: number;\n      finishedAt?: number;\n    };\n  };\n  todo_write: {\n    todos: ({\n      content: string;\n      status: \"pending\" | \"in_progress\" | \"completed\";\n    })[];\n    counts: {\n      pending: number;\n      inProgress: number;\n      completed: number;\n    };\n  };\n  workflow: {\n    runId: string;\n    agentsStarted: number;\n    result: JsonValue;\n  };\n  write: {\n    path: string;\n    operation: \"create\" | \"update\";\n    before: string | null;\n    after: string;\n  };\n}\n\ntype ToolName = keyof ToolOutputMap\n\ndeclare class ToolCallError extends Error {\n  readonly name: \"ToolCallError\";\n  readonly toolName: ToolName;\n}\n\ndeclare const tools: {\n  [K in ToolName]: (args: ToolArgsMap[K]) => Promise<ToolOutputMap[K]>;\n}\n```","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under <mode> mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_inspect","description":"Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc.","parameters":{"type":"object","properties":{"what":{"type":"string","description":"Limit the report to one section. Omit for all sections.","enum":["services","plugins","tools","temporary","api","events"]},"name":{"type":"string","description":"Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\"."}}}},{"name":"cordis_mount","description":"Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:<id>]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"JavaScript body returning a temporary Plugin; evaluated now and saved nowhere."}},"required":["code"]}},{"name":"cordis_unmount","description":"Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins.","parameters":{"type":"object","properties":{"id":{"type":"string","description":"The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart."}},"required":["id"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"run_code","description":"Execute a TypeScript program against the available tools. Write the BODY of an async function (erasable syntax only; top-level `await` and `return` work) and call tools as `await tools.name(args)` per the declarations in the system prompt. Only what you print or return comes back — curate it.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"The program: the body of an async TypeScript function."},"description":{"type":"string","description":"Clear, concise description of what this program does in active voice, 5-10 words (shown in the UI). Examples: \"Count TODO markers across packages\"; \"Read failing test and its fixture\"; \"Rename config key in every cordis.yml\"."}},"required":["code","description"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return <value>` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise<any>` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise<any[]>` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise<any[]>` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return <json-value>`)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}}
 {"type":"request/context","seq":7,"time":1785730501403,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}}
 {"type":"assistant/chunk","seq":8,"time":1783950000007,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}
 {"type":"assistant/chunk","seq":9,"time":1783950000008,"data":{"turn":1,"step":1,"chunk":{"type":"tool-call-delta","index":0,"id":"advanced-mount","name":"cordis_mount","argumentsDelta":"{\"code\":\"return { name: 'snapshot-marker', apply() {} }\"}"}}}
 {"type":"assistant/chunk","seq":10,"time":1783950000009,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"advanced-mount","name":"cordis_mount","arguments":"{\"code\":\"return { name: 'snapshot-marker', apply() {} }\"}"}}}}
 {"type":"assistant/chunk","seq":11,"time":1785498583784,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}
 {"type":"assistant/chunk","seq":12,"time":1785730501404,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
-{"type":"assistant/message","seq":13,"time":1785730501404,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-mount","name":"cordis_mount","arguments":"{\"code\":\"return { name: 'snapshot-marker', apply() {} }\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"b596133f-aafe-4485-9871-ade1dda23373"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[8,9,10,11,12],"surfaceOp":"append"}
+{"type":"assistant/message","seq":13,"time":1785730501404,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-mount","name":"cordis_mount","arguments":"{\"code\":\"return { name: 'snapshot-marker', apply() {} }\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"b4b93190-da1c-48af-aa9d-9a141da54305"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[8,9,10,11,12],"surfaceOp":"append"}
 {"type":"tool/call","seq":14,"time":1785730501404,"data":{"turn":1,"step":1,"callId":"advanced-mount","name":"cordis_mount","arguments":"{\"code\":\"return { name: 'snapshot-marker', apply() {} }\"}"}}
-{"type":"tool/result","seq":15,"time":1785730501413,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"advanced-mount"},"content":[{"type":"tool-result","toolCallId":"advanced-mount","content":[{"type":"text","text":"Temporary Plugin dyn-1 is running (plugin \"snapshot-marker\"; available until unmounted or DSH restarts)."}],"isError":false}],"role":"user","id":"99819f4a-5e53-4a6f-92f6-5be96b765bce"}},"sourceEventSeqs":[14],"surfaceOp":"append"}
+{"type":"tool/result","seq":15,"time":1785730501413,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"advanced-mount"},"content":[{"type":"tool-result","toolCallId":"advanced-mount","content":[{"type":"text","text":"Temporary Plugin dyn-1 is running (plugin \"snapshot-marker\"; available until unmounted or DSH restarts)."}],"isError":false}],"role":"user","id":"3649a228-ac4b-4539-83c6-71ec8e73da32"}},"sourceEventSeqs":[14],"surfaceOp":"append"}
 {"type":"step/end","seq":16,"time":1785730501413,"data":{"turn":1,"step":1}}
 {"type":"step/start","seq":17,"time":1785730501423,"data":{"turn":1,"step":2}}
 {"type":"assistant/chunk","seq":18,"time":1783950000017,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}
@@ -22,11 +22,11 @@
 {"type":"assistant/chunk","seq":20,"time":1783950000019,"data":{"turn":1,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"advanced-code","name":"run_code","arguments":"{\"code\": \"return await tools.cordis_inspect({ what: 'temporary' })\", \"description\": \"Run the scripted inspection program\"}"}}}}
 {"type":"assistant/chunk","seq":21,"time":1785498583804,"data":{"turn":1,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}
 {"type":"assistant/chunk","seq":22,"time":1785730501424,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
-{"type":"assistant/message","seq":23,"time":1785730501424,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-code","name":"run_code","arguments":"{\"code\": \"return await tools.cordis_inspect({ what: 'temporary' })\", \"description\": \"Run the scripted inspection program\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"6697e967-e6bd-46b2-8574-18aeb914e7c6"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[18,19,20,21,22],"surfaceOp":"append"}
+{"type":"assistant/message","seq":23,"time":1785730501424,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-code","name":"run_code","arguments":"{\"code\": \"return await tools.cordis_inspect({ what: 'temporary' })\", \"description\": \"Run the scripted inspection program\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"b216bd0d-8dc7-4f88-951b-db1544e20503"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[18,19,20,21,22],"surfaceOp":"append"}
 {"type":"tool/call","seq":24,"time":1785730501424,"data":{"turn":1,"step":2,"callId":"advanced-code","name":"run_code","arguments":"{\"code\": \"return await tools.cordis_inspect({ what: 'temporary' })\", \"description\": \"Run the scripted inspection program\"}"}}
 {"type":"tool/code-dispatch-start","seq":25,"time":1785730501473,"data":{"rootCallId":"advanced-code","parentCallId":"advanced-code","subCallId":"advanced-code:code:1","name":"cordis_inspect","arguments":{"what":"temporary"}}}
 {"type":"tool/code-dispatch","seq":26,"time":1785730501474,"data":{"rootCallId":"advanced-code","parentCallId":"advanced-code","subCallId":"advanced-code:code:1","name":"cordis_inspect","arguments":{"what":"temporary"},"isError":false,"content":[{"type":"text","text":"## Temporary Plugins\n- Temporary Plugin dyn-1: snapshot-marker [running] — provides: none; waiting for: none; lifetime: until unmounted or DSH restarts"}]}}
-{"type":"tool/result","seq":27,"time":1785730501475,"data":{"turn":1,"step":2,"message":{"source":{"kind":"tool","callId":"advanced-code"},"content":[{"type":"tool-result","toolCallId":"advanced-code","content":[{"type":"text","text":"## Temporary Plugins\n- Temporary Plugin dyn-1: snapshot-marker [running] — provides: none; waiting for: none; lifetime: until unmounted or DSH restarts"}],"isError":false}],"role":"user","id":"831070a8-3dc0-4275-9f11-0476b47b8ef2"}},"sourceEventSeqs":[24],"surfaceOp":"append"}
+{"type":"tool/result","seq":27,"time":1785730501475,"data":{"turn":1,"step":2,"message":{"source":{"kind":"tool","callId":"advanced-code"},"content":[{"type":"tool-result","toolCallId":"advanced-code","content":[{"type":"text","text":"## Temporary Plugins\n- Temporary Plugin dyn-1: snapshot-marker [running] — provides: none; waiting for: none; lifetime: until unmounted or DSH restarts"}],"isError":false}],"role":"user","id":"c27e4cc6-d4fd-414a-83f1-20962a03f00f"}},"sourceEventSeqs":[24],"surfaceOp":"append"}
 {"type":"step/end","seq":28,"time":1785730501475,"data":{"turn":1,"step":2}}
 {"type":"step/start","seq":29,"time":1785730501483,"data":{"turn":1,"step":3}}
 {"type":"assistant/chunk","seq":30,"time":1783950000029,"data":{"turn":1,"step":3,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}
@@ -34,9 +34,9 @@
 {"type":"assistant/chunk","seq":32,"time":1785037378923,"data":{"turn":1,"step":3,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"advanced-direct-child","name":"subagent","arguments":"{\"description\":\"Check direct child\",\"prompt\":\"Reply with exactly DIRECT_CHILD_OK and nothing else.\"}"}}}}
 {"type":"assistant/chunk","seq":33,"time":1785498583869,"data":{"turn":1,"step":3,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}
 {"type":"assistant/chunk","seq":34,"time":1785730501484,"data":{"turn":1,"step":3,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
-{"type":"assistant/message","seq":35,"time":1785730501484,"data":{"turn":1,"step":3,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-direct-child","name":"subagent","arguments":"{\"description\":\"Check direct child\",\"prompt\":\"Reply with exactly DIRECT_CHILD_OK and nothing else.\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"cef24cf0-5f9e-4be2-93d8-93a0d89c6e82"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[30,31,32,33,34],"surfaceOp":"append"}
+{"type":"assistant/message","seq":35,"time":1785730501484,"data":{"turn":1,"step":3,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-direct-child","name":"subagent","arguments":"{\"description\":\"Check direct child\",\"prompt\":\"Reply with exactly DIRECT_CHILD_OK and nothing else.\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"519d60a3-3d6a-4a21-830d-fe023de72969"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[30,31,32,33,34],"surfaceOp":"append"}
 {"type":"tool/call","seq":36,"time":1785730501484,"data":{"turn":1,"step":3,"callId":"advanced-direct-child","name":"subagent","arguments":"{\"description\":\"Check direct child\",\"prompt\":\"Reply with exactly DIRECT_CHILD_OK and nothing else.\"}"}}
-{"type":"tool/result","seq":37,"time":1785730501508,"data":{"turn":1,"step":3,"message":{"source":{"kind":"tool","callId":"advanced-direct-child"},"content":[{"type":"tool-result","toolCallId":"advanced-direct-child","content":[{"type":"text","text":"DIRECT_CHILD_OK"}],"isError":false}],"role":"user","id":"d7ea395e-840a-46c4-a143-b6e15cf74114"}},"sourceEventSeqs":[36],"surfaceOp":"append"}
+{"type":"tool/result","seq":37,"time":1785730501508,"data":{"turn":1,"step":3,"message":{"source":{"kind":"tool","callId":"advanced-direct-child"},"content":[{"type":"tool-result","toolCallId":"advanced-direct-child","content":[{"type":"text","text":"DIRECT_CHILD_OK"}],"isError":false}],"role":"user","id":"a7f38fae-607d-4b35-a213-7cd59027c6d2"}},"sourceEventSeqs":[36],"surfaceOp":"append"}
 {"type":"step/end","seq":38,"time":1785730501508,"data":{"turn":1,"step":3}}
 {"type":"step/start","seq":39,"time":1785730501521,"data":{"turn":1,"step":4}}
 {"type":"assistant/chunk","seq":40,"time":1783957884594,"data":{"turn":1,"step":4,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}
@@ -44,9 +44,9 @@
 {"type":"assistant/chunk","seq":42,"time":1785037378946,"data":{"turn":1,"step":4,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"advanced-workflow","name":"workflow","arguments":"{\"script\":\"phase('Delegate')\\nconst reply = await agent('Reply with exactly WORKFLOW_CHILD_OK and nothing else.', { label: 'workflow-child' })\\nreturn { reply }\",\"meta\":{\"name\":\"advanced-headless-snapshot\",\"description\":\"exercise one workflow child through the headless agent\"}}"}}}}
 {"type":"assistant/chunk","seq":43,"time":1785498583919,"data":{"turn":1,"step":4,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}
 {"type":"assistant/chunk","seq":44,"time":1785730501522,"data":{"turn":1,"step":4,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
-{"type":"assistant/message","seq":45,"time":1785730501522,"data":{"turn":1,"step":4,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-workflow","name":"workflow","arguments":"{\"script\":\"phase('Delegate')\\nconst reply = await agent('Reply with exactly WORKFLOW_CHILD_OK and nothing else.', { label: 'workflow-child' })\\nreturn { reply }\",\"meta\":{\"name\":\"advanced-headless-snapshot\",\"description\":\"exercise one workflow child through the headless agent\"}}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"242be4fa-3293-45de-ab55-a017999f2333"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[40,41,42,43,44],"surfaceOp":"append"}
+{"type":"assistant/message","seq":45,"time":1785730501522,"data":{"turn":1,"step":4,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-workflow","name":"workflow","arguments":"{\"script\":\"phase('Delegate')\\nconst reply = await agent('Reply with exactly WORKFLOW_CHILD_OK and nothing else.', { label: 'workflow-child' })\\nreturn { reply }\",\"meta\":{\"name\":\"advanced-headless-snapshot\",\"description\":\"exercise one workflow child through the headless agent\"}}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"ff213e7e-70da-475b-ae45-0c2149ad73be"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[40,41,42,43,44],"surfaceOp":"append"}
 {"type":"tool/call","seq":46,"time":1785730501522,"data":{"turn":1,"step":4,"callId":"advanced-workflow","name":"workflow","arguments":"{\"script\":\"phase('Delegate')\\nconst reply = await agent('Reply with exactly WORKFLOW_CHILD_OK and nothing else.', { label: 'workflow-child' })\\nreturn { reply }\",\"meta\":{\"name\":\"advanced-headless-snapshot\",\"description\":\"exercise one workflow child through the headless agent\"}}"}}
-{"type":"tool/result","seq":47,"time":1785730501647,"data":{"turn":1,"step":4,"message":{"source":{"kind":"tool","callId":"advanced-workflow"},"content":[{"type":"tool-result","toolCallId":"advanced-workflow","content":[{"type":"text","text":"workflow \"advanced-headless-snapshot\" completed (1 agent).\nReturn value:\n{\n  \"reply\": \"WORKFLOW_CHILD_OK\"\n}"}],"isError":false}],"role":"user","id":"ee764f11-8827-4984-acef-ec3c5880f5a0"}},"sourceEventSeqs":[46],"surfaceOp":"append"}
+{"type":"tool/result","seq":47,"time":1785730501647,"data":{"turn":1,"step":4,"message":{"source":{"kind":"tool","callId":"advanced-workflow"},"content":[{"type":"tool-result","toolCallId":"advanced-workflow","content":[{"type":"text","text":"workflow \"advanced-headless-snapshot\" completed (1 agent).\nReturn value:\n{\n  \"reply\": \"WORKFLOW_CHILD_OK\"\n}"}],"isError":false}],"role":"user","id":"49733704-4123-47d8-9ea4-ac4f4cf84362"}},"sourceEventSeqs":[46],"surfaceOp":"append"}
 {"type":"step/end","seq":48,"time":1785730501648,"data":{"turn":1,"step":4}}
 {"type":"step/start","seq":49,"time":1785730501660,"data":{"turn":1,"step":5}}
 {"type":"assistant/chunk","seq":50,"time":1783957884719,"data":{"turn":1,"step":5,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}
@@ -54,9 +54,9 @@
 {"type":"assistant/chunk","seq":52,"time":1785037379534,"data":{"turn":1,"step":5,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"advanced-unmount","name":"cordis_unmount","arguments":"{\"id\":\"dyn-1\"}"}}}}
 {"type":"assistant/chunk","seq":53,"time":1785498584085,"data":{"turn":1,"step":5,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}
 {"type":"assistant/chunk","seq":54,"time":1785730501661,"data":{"turn":1,"step":5,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
-{"type":"assistant/message","seq":55,"time":1785730501661,"data":{"turn":1,"step":5,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-unmount","name":"cordis_unmount","arguments":"{\"id\":\"dyn-1\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"6c3dab82-c2ed-492a-ab0d-f235b340a6c1"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[50,51,52,53,54],"surfaceOp":"append"}
+{"type":"assistant/message","seq":55,"time":1785730501661,"data":{"turn":1,"step":5,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-unmount","name":"cordis_unmount","arguments":"{\"id\":\"dyn-1\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"afb985f3-c44a-42fe-affa-4d87adf8311c"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[50,51,52,53,54],"surfaceOp":"append"}
 {"type":"tool/call","seq":56,"time":1785730501661,"data":{"turn":1,"step":5,"callId":"advanced-unmount","name":"cordis_unmount","arguments":"{\"id\":\"dyn-1\"}"}}
-{"type":"tool/result","seq":57,"time":1785730501668,"data":{"turn":1,"step":5,"message":{"source":{"kind":"tool","callId":"advanced-unmount"},"content":[{"type":"tool-result","toolCallId":"advanced-unmount","content":[{"type":"text","text":"Temporary Plugin dyn-1 was unmounted and removed."}],"isError":false}],"role":"user","id":"74554b42-640e-4119-a413-ee5ac01e546e"}},"sourceEventSeqs":[56],"surfaceOp":"append"}
+{"type":"tool/result","seq":57,"time":1785730501668,"data":{"turn":1,"step":5,"message":{"source":{"kind":"tool","callId":"advanced-unmount"},"content":[{"type":"tool-result","toolCallId":"advanced-unmount","content":[{"type":"text","text":"Temporary Plugin dyn-1 was unmounted and removed."}],"isError":false}],"role":"user","id":"6ba07bf6-fe0b-4ba2-ac1e-f01fc5da7120"}},"sourceEventSeqs":[56],"surfaceOp":"append"}
 {"type":"step/end","seq":58,"time":1785730501668,"data":{"turn":1,"step":5}}
 {"type":"step/start","seq":59,"time":1785730501678,"data":{"turn":1,"step":6}}
 {"type":"assistant/chunk","seq":60,"time":1783957884720,"data":{"turn":1,"step":6,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
@@ -64,6 +64,6 @@
 {"type":"assistant/chunk","seq":62,"time":1785037379541,"data":{"turn":1,"step":6,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"ADVANCED_HEADLESS_OK"}}}}
 {"type":"assistant/chunk","seq":63,"time":1785498584102,"data":{"turn":1,"step":6,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}
 {"type":"assistant/chunk","seq":64,"time":1785730501679,"data":{"turn":1,"step":6,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}
-{"type":"assistant/message","seq":65,"time":1785730501679,"data":{"turn":1,"step":6,"message":{"role":"assistant","content":[{"type":"text","text":"ADVANCED_HEADLESS_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"1a3bef32-0610-4891-9071-6bdc2e8a8fd2"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[60,61,62,63,64],"surfaceOp":"append"}
+{"type":"assistant/message","seq":65,"time":1785730501679,"data":{"turn":1,"step":6,"message":{"role":"assistant","content":[{"type":"text","text":"ADVANCED_HEADLESS_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"067c0430-9c91-46d8-8de5-fd58c606a8a2"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[60,61,62,63,64],"surfaceOp":"append"}
 {"type":"step/end","seq":66,"time":1785730501679,"data":{"turn":1,"step":6}}
 {"type":"turn/end","seq":67,"time":1785730501679,"data":{"turn":1,"reason":{"kind":"completed"}}}
diff --git a/examples/headless-agent/tests/snapshots/compaction-recovery/session.jsonl b/examples/headless-agent/tests/snapshots/compaction-recovery/session.jsonl
index 9e39f76b11..04f5f3e788 100644
--- a/examples/headless-agent/tests/snapshots/compaction-recovery/session.jsonl
+++ b/examples/headless-agent/tests/snapshots/compaction-recovery/session.jsonl
@@ -1,32 +1,32 @@
 {"type":"session","version":0,"id":"{{sessionId}}","createdAt":0,"cwd":"{{cwd}}","delegationDepth":0}
-{"type":"agent/inbox/spliced","seq":0,"time":1786123401613,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Establish a durable compaction premise before continuing. Record every part of this historical evidence: the snapshot uses keyless replay; persistence uses JSONL; the assembled headless application loads its real Cordis composition; model-visible inputs remain logged; tool calls and results remain paired and ordered; context overflow retains the original failure while recovery is attempted; compaction opens with compact/start and closes with compact/end; a successful auxiliary summary records compact/summary provenance; the replacement surface shadows only an older balanced range; the checkpoint remains smaller than the history it replaces; the newest tool result remains verbatim; the retried request sees that checkpoint; the final response proves the same turn continued; deterministic snapshot evidence stays separate from the live-provider smoke; and no external API key is needed. Emit one alpha marker through bash, then finish the task after any required recovery with the exact words COMPACTION RECOVERED."}],"source":{"kind":"user"},"role":"user","id":"203f911d-905a-43e4-8e1c-fccfe78d445b"}]}}
+{"type":"agent/inbox/spliced","seq":0,"time":1786123401613,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Establish a durable compaction premise before continuing. Record every part of this historical evidence: the snapshot uses keyless replay; persistence uses JSONL; the assembled headless application loads its real Cordis composition; model-visible inputs remain logged; tool calls and results remain paired and ordered; context overflow retains the original failure while recovery is attempted; compaction opens with compact/start and closes with compact/end; a successful auxiliary summary records compact/summary provenance; the replacement surface shadows only an older balanced range; the checkpoint remains smaller than the history it replaces; the newest tool result remains verbatim; the retried request sees that checkpoint; the final response proves the same turn continued; deterministic snapshot evidence stays separate from the live-provider smoke; and no external API key is needed. Emit one alpha marker through bash, then finish the task after any required recovery with the exact words COMPACTION RECOVERED."}],"source":{"kind":"user"},"role":"user","id":"01b4f117-d618-45e0-af9b-7db6e054dcfe"}]}}
 {"type":"turn/start","seq":1,"time":1786123401614,"data":{"turn":1}}
 {"type":"agent/inbox/spliced","seq":2,"time":1786123401614,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}
 {"type":"step/start","seq":3,"time":1786123401667,"data":{"turn":1,"step":1}}
-{"type":"user/message","seq":4,"time":1786123401667,"data":{"content":[{"type":"text","text":"Establish a durable compaction premise before continuing. Record every part of this historical evidence: the snapshot uses keyless replay; persistence uses JSONL; the assembled headless application loads its real Cordis composition; model-visible inputs remain logged; tool calls and results remain paired and ordered; context overflow retains the original failure while recovery is attempted; compaction opens with compact/start and closes with compact/end; a successful auxiliary summary records compact/summary provenance; the replacement surface shadows only an older balanced range; the checkpoint remains smaller than the history it replaces; the newest tool result remains verbatim; the retried request sees that checkpoint; the final response proves the same turn continued; deterministic snapshot evidence stays separate from the live-provider smoke; and no external API key is needed. Emit one alpha marker through bash, then finish the task after any required recovery with the exact words COMPACTION RECOVERED."}],"source":{"kind":"user"},"role":"user","id":"203f911d-905a-43e4-8e1c-fccfe78d445b"},"surfaceOp":"append"}
+{"type":"user/message","seq":4,"time":1786123401667,"data":{"content":[{"type":"text","text":"Establish a durable compaction premise before continuing. Record every part of this historical evidence: the snapshot uses keyless replay; persistence uses JSONL; the assembled headless application loads its real Cordis composition; model-visible inputs remain logged; tool calls and results remain paired and ordered; context overflow retains the original failure while recovery is attempted; compaction opens with compact/start and closes with compact/end; a successful auxiliary summary records compact/summary provenance; the replacement surface shadows only an older balanced range; the checkpoint remains smaller than the history it replaces; the newest tool result remains verbatim; the retried request sees that checkpoint; the final response proves the same turn continued; deterministic snapshot evidence stays separate from the live-provider smoke; and no external API key is needed. Emit one alpha marker through bash, then finish the task after any required recovery with the exact words COMPACTION RECOVERED."}],"source":{"kind":"user"},"role":"user","id":"01b4f117-d618-45e0-af9b-7db6e054dcfe"},"surfaceOp":"append"}
 {"type":"session/title","seq":5,"time":1786123401667,"data":{"title":"Establish a durable compaction premise","messageSeqs":[4],"source":{"kind":"fallback"}}}
-{"type":"request/header","seq":6,"time":1786123401668,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model.\n\nVerify your work by running the code or tests. Keep answers brief and\nfactual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under <mode> mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return <value>` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise<any>` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise<any[]>` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise<any[]>` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return <json-value>`)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}}
+{"type":"request/header","seq":6,"time":1786123401668,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model.\n\nVerify your work by running the code or tests. Keep answers brief and\nfactual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under <mode> mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return <value>` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise<any>` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise<any[]>` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise<any[]>` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return <json-value>`)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}}
 {"type":"request/context","seq":7,"time":1786123401669,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash","contextWindow":128000}}
 {"type":"assistant/chunk","seq":8,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}
 {"type":"assistant/chunk","seq":9,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"tool-call-delta","index":0,"id":"call_compaction_marker","name":"bash","argumentsDelta":"{\"command\":\"printf 'alpha\\n'\",\"description\":\"Emit compaction premise marker\"}"}}}
 {"type":"assistant/chunk","seq":10,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"call_compaction_marker","name":"bash","arguments":"{\"command\":\"printf 'alpha\\n'\",\"description\":\"Emit compaction premise marker\"}"}}}}
 {"type":"assistant/chunk","seq":11,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":24,"outputTokens":6}}}}
 {"type":"assistant/chunk","seq":12,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
-{"type":"assistant/message","seq":13,"time":1786123401680,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_compaction_marker","name":"bash","arguments":"{\"command\":\"printf 'alpha\\n'\",\"description\":\"Emit compaction premise marker\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"6e7dfb28-f033-4dbf-9fd9-6b2f71da6906"},"usage":{"inputTokens":24,"outputTokens":6}},"sourceEventSeqs":[8,9,10,11,12],"surfaceOp":"append"}
+{"type":"assistant/message","seq":13,"time":1786123401680,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_compaction_marker","name":"bash","arguments":"{\"command\":\"printf 'alpha\\n'\",\"description\":\"Emit compaction premise marker\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"edc2376e-f087-4a28-9dce-70b204457a7c"},"usage":{"inputTokens":24,"outputTokens":6}},"sourceEventSeqs":[8,9,10,11,12],"surfaceOp":"append"}
 {"type":"tool/call","seq":14,"time":1786123401680,"data":{"turn":1,"step":1,"callId":"call_compaction_marker","name":"bash","arguments":"{\"command\":\"printf 'alpha\\n'\",\"description\":\"Emit compaction premise marker\"}"}}
-{"type":"tool/result","seq":15,"time":1786123401700,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_compaction_marker"},"content":[{"type":"tool-result","toolCallId":"call_compaction_marker","content":[{"type":"text","text":"alpha\n"}],"isError":false}],"role":"user","id":"4fe4e523-3ac9-421c-8a73-1eb033ac8b09"}},"sourceEventSeqs":[14],"surfaceOp":"append"}
+{"type":"tool/result","seq":15,"time":1786123401700,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_compaction_marker"},"content":[{"type":"tool-result","toolCallId":"call_compaction_marker","content":[{"type":"text","text":"alpha\n"}],"isError":false}],"role":"user","id":"fc210d7a-92aa-49fc-ab4c-fc5ddbf5a756"}},"sourceEventSeqs":[14],"surfaceOp":"append"}
 {"type":"step/end","seq":16,"time":1786123401700,"data":{"turn":1,"step":1}}
 {"type":"step/start","seq":17,"time":1786123401710,"data":{"turn":1,"step":2}}
 {"type":"assistant/chunk","seq":18,"time":1786123401715,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"error","failure":{"message":"snapshot request exceeded the model context window","code":"CONTEXT_WINDOW_EXCEEDED"}}}}}
 {"type":"compact/start","seq":19,"time":1786123401715,"data":{"compactionId":"338e88fa-e78b-4d49-bd38-8f919e85f5e1","turn":1}}
 {"type":"compact/summary","seq":20,"time":1786123401725,"data":{"compactionId":"338e88fa-e78b-4d49-bd38-8f919e85f5e1","summary":[{"type":"text","text":"The request established a durable compaction premise."}],"rawOutput":[{"type":"text","text":"The request established a durable compaction premise."}],"llmStreamCall":true,"shadowedRange":{"start":4,"end":4},"shadowedSeqs":[4],"shadowedTokenCount":264,"provider":"deepseek-official","model":"deepseek-v4-flash","maxTokens":32,"usage":{"inputTokens":20,"outputTokens":4}}}
-{"type":"user/message","seq":21,"time":1786123401725,"data":{"content":[{"type":"text","text":"This is an automatically generated checkpoint condensing an earlier span of the conversation to free up context. Treat the captured context as established background and build on it without restating it. Continue the task directly from the messages that follow, without acknowledging this checkpoint.\n\n<compacted-summary>"},{"type":"text","text":"The request established a durable compaction premise."},{"type":"text","text":"</compacted-summary>"}],"source":{"kind":"plugin","plugin":"compact","compactionId":"338e88fa-e78b-4d49-bd38-8f919e85f5e1"},"role":"user","id":"b74220bf-4104-4c30-b967-e4570d5eba5d"},"sourceEventSeqs":[19,20,4],"surfaceOp":{"op":"replace","start":4,"end":4}}
+{"type":"user/message","seq":21,"time":1786123401725,"data":{"content":[{"type":"text","text":"This is an automatically generated checkpoint condensing an earlier span of the conversation to free up context. Treat the captured context as established background and build on it without restating it. Continue the task directly from the messages that follow, without acknowledging this checkpoint.\n\n<compacted-summary>"},{"type":"text","text":"The request established a durable compaction premise."},{"type":"text","text":"</compacted-summary>"}],"source":{"kind":"plugin","plugin":"compact","compactionId":"338e88fa-e78b-4d49-bd38-8f919e85f5e1"},"role":"user","id":"1f0c6a76-d3e1-4776-9d41-484ce58a49c9"},"sourceEventSeqs":[19,20,4],"surfaceOp":{"op":"replace","start":4,"end":4}}
 {"type":"compact/end","seq":22,"time":1786123401725,"data":{"compactionId":"338e88fa-e78b-4d49-bd38-8f919e85f5e1","turn":1}}
 {"type":"assistant/chunk","seq":23,"time":1786123401730,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
 {"type":"assistant/chunk","seq":24,"time":1786123401730,"data":{"turn":1,"step":2,"chunk":{"type":"text-delta","index":0,"text":"COMPACTION RECOVERED"}}}
 {"type":"assistant/chunk","seq":25,"time":1786123401730,"data":{"turn":1,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"COMPACTION RECOVERED"}}}}
 {"type":"assistant/chunk","seq":26,"time":1786123401730,"data":{"turn":1,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":20,"outputTokens":4}}}}
 {"type":"assistant/chunk","seq":27,"time":1786123401730,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}
-{"type":"assistant/message","seq":28,"time":1786123401730,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"text","text":"COMPACTION RECOVERED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"103034c2-e156-4808-ba2e-8e17e7107c9e"},"usage":{"inputTokens":20,"outputTokens":4}},"sourceEventSeqs":[23,24,25,26,27],"surfaceOp":"append"}
+{"type":"assistant/message","seq":28,"time":1786123401730,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"text","text":"COMPACTION RECOVERED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"be904dc2-41ce-4df8-9308-3cdd81d6eb53"},"usage":{"inputTokens":20,"outputTokens":4}},"sourceEventSeqs":[23,24,25,26,27],"surfaceOp":"append"}
 {"type":"step/end","seq":29,"time":1786123401730,"data":{"turn":1,"step":2}}
 {"type":"turn/end","seq":30,"time":1786123401730,"data":{"turn":1,"reason":{"kind":"completed"}}}
diff --git a/examples/headless-agent/tests/snapshots/pty-tools/session.jsonl b/examples/headless-agent/tests/snapshots/pty-tools/session.jsonl
index f4f592570a..a6037bdc27 100644
--- a/examples/headless-agent/tests/snapshots/pty-tools/session.jsonl
+++ b/examples/headless-agent/tests/snapshots/pty-tools/session.jsonl
@@ -1,21 +1,21 @@
 {"type":"session","version":0,"id":"{{sessionId}}","createdAt":0,"cwd":"{{cwd}}","delegationDepth":0}
-{"type":"agent/inbox/spliced","seq":0,"time":1785498587408,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Exercise the six PTY tools in order, including one missing-session signal error, then reply DONE."}],"source":{"kind":"user"},"role":"user","id":"39f39ecc-5772-4814-8feb-46433c71becd"}]}}
+{"type":"agent/inbox/spliced","seq":0,"time":1785498587408,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Exercise the six PTY tools in order, including one missing-session signal error, then reply DONE."}],"source":{"kind":"user"},"role":"user","id":"a8faabbd-30ed-4322-9233-102a81aec70c"}]}}
 {"type":"turn/start","seq":1,"time":1785821457966,"data":{"turn":1}}
 {"type":"agent/inbox/spliced","seq":2,"time":1785821457966,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}
 {"type":"step/start","seq":3,"time":0,"data":{"turn":1,"step":1}}
-{"type":"user/message","seq":4,"time":1785498587436,"data":{"content":[{"type":"text","text":"Exercise the six PTY tools in order, including one missing-session signal error, then reply DONE."}],"source":{"kind":"user"},"role":"user","id":"39f39ecc-5772-4814-8feb-46433c71becd"},"surfaceOp":"append"}
-{"type":"user/message","seq":5,"time":1785730504659,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."}]},"role":"user","id":"a5ae9c04-0652-436f-9b5a-437a3a6ed235"},"surfaceOp":"append"}
+{"type":"user/message","seq":4,"time":1785498587436,"data":{"content":[{"type":"text","text":"Exercise the six PTY tools in order, including one missing-session signal error, then reply DONE."}],"source":{"kind":"user"},"role":"user","id":"a8faabbd-30ed-4322-9233-102a81aec70c"},"surfaceOp":"append"}
+{"type":"user/message","seq":5,"time":1785730504659,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."}]},"role":"user","id":"a3fb63b8-f264-4fc7-a745-52ea26e3768a"},"surfaceOp":"append"}
 {"type":"session/title","seq":6,"time":1785730504659,"data":{"title":"Exercise the six PTY tools","messageSeqs":[4],"source":{"kind":"fallback"}}}
-{"type":"request/header","seq":7,"time":1785498587438,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model.\n\nVerify your work by running the code or tests. Keep answers brief and\nfactual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nUse a terminal session only when work needs persistent terminal state or interactive stdin; prefer bash/read/write/edit for bounded one-shot operations. Track every terminal session id and close sessions that no longer matter. An inferred_idle or timeout result does not prove the foreground command exited.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under <mode> mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"terminal_close","description":"Close one persistent terminal and wait until its captured owned process tree is gone.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id."}},"required":["sessionId"]}},{"name":"terminal_list","description":"List persistent terminal sessions owned by the current agent.","parameters":{"type":"object","properties":{}}},{"name":"terminal_open","description":"Create a persistent, owner-isolated terminal session from a registered backend type. Use this for shell or REPL state that must survive across tool calls.","parameters":{"type":"object","properties":{"type":{"type":"string","description":"Registered terminal backend type, usually \"shell\"."},"name":{"type":"string","description":"Optional owner-local display name such as \"main\" or \"gdb\"."},"cwd":{"type":"string","description":"Initial working directory. Defaults to the deployment workspace root."}},"required":["type"]}},{"name":"terminal_read","description":"Read a bounded page of retained output from a persistent terminal without sending input.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id."},"offset":{"type":"number","description":"Newest-relative line offset (default 0)."},"count":{"type":"number","description":"Requested line count (default 500; backend caps apply)."}},"required":["sessionId"]}},{"name":"terminal_send","description":"Send text to a persistent terminal. By default Enter is submitted and the call waits for a prompt, stdin wait, output silence, timeout, or session exit. Background mode returns a task id for task_output/task_kill.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id returned by terminal_open or terminal_list."},"text":{"type":"string","description":"UTF-8 text to write to the terminal."},"submit":{"type":"boolean","description":"Submit Enter after text (default true). Set false for control characters or incomplete REPL input."},"run_in_background":{"type":"boolean","description":"Return a task id immediately; collect with task_output or stop with task_kill."}},"required":["sessionId","text"]}},{"name":"terminal_signal","description":"Send an allowed signal to the current foreground process group of a persistent terminal.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id."},"signal":{"type":"string","description":"Signal to deliver. Shell-targeted SIGKILL is rejected; use terminal_close.","enum":["SIGINT","SIGTERM","SIGKILL","SIGTSTP","SIGHUP"]}},"required":["sessionId","signal"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return <value>` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise<any>` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise<any[]>` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise<any[]>` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return <json-value>`)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}}
+{"type":"request/header","seq":7,"time":1785498587438,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model.\n\nVerify your work by running the code or tests. Keep answers brief and\nfactual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nUse a terminal session only when work needs persistent terminal state or interactive stdin; prefer bash/read/write/edit for bounded one-shot operations. Track every terminal session id and close sessions that no longer matter. An inferred_idle or timeout result does not prove the foreground command exited.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under <mode> mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"terminal_close","description":"Close one persistent terminal and wait until its captured owned process tree is gone.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id."}},"required":["sessionId"]}},{"name":"terminal_list","description":"List persistent terminal sessions owned by the current agent.","parameters":{"type":"object","properties":{}}},{"name":"terminal_open","description":"Create a persistent, owner-isolated terminal session from a registered backend type. Use this for shell or REPL state that must survive across tool calls.","parameters":{"type":"object","properties":{"type":{"type":"string","description":"Registered terminal backend type, usually \"shell\"."},"name":{"type":"string","description":"Optional owner-local display name such as \"main\" or \"gdb\"."},"cwd":{"type":"string","description":"Initial working directory. Defaults to the deployment workspace root."}},"required":["type"]}},{"name":"terminal_read","description":"Read a bounded page of retained output from a persistent terminal without sending input.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id."},"offset":{"type":"number","description":"Newest-relative line offset (default 0)."},"count":{"type":"number","description":"Requested line count (default 500; backend caps apply)."}},"required":["sessionId"]}},{"name":"terminal_send","description":"Send text to a persistent terminal. By default Enter is submitted and the call waits for a prompt, stdin wait, output silence, timeout, or session exit. Background mode returns a task id for task_output/task_kill.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id returned by terminal_open or terminal_list."},"text":{"type":"string","description":"UTF-8 text to write to the terminal."},"submit":{"type":"boolean","description":"Submit Enter after text (default true). Set false for control characters or incomplete REPL input."},"run_in_background":{"type":"boolean","description":"Return a task id immediately; collect with task_output or stop with task_kill."}},"required":["sessionId","text"]}},{"name":"terminal_signal","description":"Send an allowed signal to the current foreground process group of a persistent terminal.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id."},"signal":{"type":"string","description":"Signal to deliver. Shell-targeted SIGKILL is rejected; use terminal_close.","enum":["SIGINT","SIGTERM","SIGKILL","SIGTSTP","SIGHUP"]}},"required":["sessionId","signal"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return <value>` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise<any>` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise<any[]>` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise<any[]>` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return <json-value>`)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}}
 {"type":"request/context","seq":8,"time":1785730504660,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}}
 {"type":"assistant/chunk","seq":9,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}
 {"type":"assistant/chunk","seq":10,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"tool-call-delta","index":0,"id":"pty-spawn","name":"terminal_open","argumentsDelta":"{\"type\":\"shell\",\"name\":\"main\"}"}}}
 {"type":"assistant/chunk","seq":11,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"pty-spawn","name":"terminal_open","arguments":"{\"type\":\"shell\",\"name\":\"main\"}"}}}}
 {"type":"assistant/chunk","seq":12,"time":1785498587439,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
 {"type":"assistant/chunk","seq":13,"time":1785730504661,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
-{"type":"assistant/message","seq":14,"time":1785730504661,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-spawn","name":"terminal_open","arguments":"{\"type\":\"shell\",\"name\":\"main\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"8cad9650-de5a-4075-8aa3-1b35e67efc2d"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[9,10,11,12,13],"surfaceOp":"append"}
+{"type":"assistant/message","seq":14,"time":1785730504661,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-spawn","name":"terminal_open","arguments":"{\"type\":\"shell\",\"name\":\"main\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"6a9b28a2-92bd-47f7-8781-33734dc85920"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[9,10,11,12,13],"surfaceOp":"append"}
 {"type":"tool/call","seq":15,"time":1785730504662,"data":{"turn":1,"step":1,"callId":"pty-spawn","name":"terminal_open","arguments":"{\"type\":\"shell\",\"name\":\"main\"}"}}
-{"type":"tool/result","seq":16,"time":1785730504671,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"pty-spawn"},"content":[{"type":"tool-result","toolCallId":"pty-spawn","content":[{"type":"text","text":"started terminal session pty-1 (main) [type: shell]\ndsh> "}],"isError":false}],"role":"user","id":"1ba1f641-0cf2-496c-895e-3982aa40b0ed"}},"sourceEventSeqs":[15],"surfaceOp":"append"}
+{"type":"tool/result","seq":16,"time":1785730504671,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"pty-spawn"},"content":[{"type":"tool-result","toolCallId":"pty-spawn","content":[{"type":"text","text":"started terminal session pty-1 (main) [type: shell]\ndsh> "}],"isError":false}],"role":"user","id":"f3229479-73c7-4c66-ac57-280d1393043a"}},"sourceEventSeqs":[15],"surfaceOp":"append"}
 {"type":"step/end","seq":17,"time":1785730504671,"data":{"turn":1,"step":1}}
 {"type":"step/start","seq":18,"time":1785730504679,"data":{"turn":1,"step":2}}
 {"type":"assistant/chunk","seq":19,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}
@@ -23,9 +23,9 @@
 {"type":"assistant/chunk","seq":21,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"pty-send","name":"terminal_send","arguments":"{\"sessionId\":\"pty-1\",\"text\":\"printf 'PTY_OK\\\\n'\"}"}}}}
 {"type":"assistant/chunk","seq":22,"time":1785498587457,"data":{"turn":1,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
 {"type":"assistant/chunk","seq":23,"time":1785730504680,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
-{"type":"assistant/message","seq":24,"time":1785730504680,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-send","name":"terminal_send","arguments":"{\"sessionId\":\"pty-1\",\"text\":\"printf 'PTY_OK\\\\n'\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"f5850535-906d-4a78-8518-a733ec91bbd8"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"}
+{"type":"assistant/message","seq":24,"time":1785730504680,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-send","name":"terminal_send","arguments":"{\"sessionId\":\"pty-1\",\"text\":\"printf 'PTY_OK\\\\n'\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"8d0e8e4c-06c7-46bc-bb3c-bfbf3806566d"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"}
 {"type":"tool/call","seq":25,"time":1785730504680,"data":{"turn":1,"step":2,"callId":"pty-send","name":"terminal_send","arguments":"{\"sessionId\":\"pty-1\",\"text\":\"printf 'PTY_OK\\\\n'\"}"}}
-{"type":"tool/result","seq":26,"time":1785730504688,"data":{"turn":1,"step":2,"message":{"source":{"kind":"tool","callId":"pty-send"},"content":[{"type":"tool-result","toolCallId":"pty-send","content":[{"type":"text","text":"K\ndsh> \n[wait: stdin_read]\n[session: running]\n[output truncated]"}],"isError":false}],"role":"user","id":"4f19b0d0-b31d-40b7-84e5-2783131cb363"},"meta":{"viewport":"printf 'PTY_OK\\n'\nPTY_OK\ndsh> ","waitReason":"stdin_read","sessionStatus":{"kind":"running"},"truncated":false}},"sourceEventSeqs":[25],"surfaceOp":"append"}
+{"type":"tool/result","seq":26,"time":1785730504688,"data":{"turn":1,"step":2,"message":{"source":{"kind":"tool","callId":"pty-send"},"content":[{"type":"tool-result","toolCallId":"pty-send","content":[{"type":"text","text":"K\ndsh> \n[wait: stdin_read]\n[session: running]\n[output truncated]"}],"isError":false}],"role":"user","id":"5d9679cb-4373-4cb2-8acd-46122e70db91"},"meta":{"viewport":"printf 'PTY_OK\\n'\nPTY_OK\ndsh> ","waitReason":"stdin_read","sessionStatus":{"kind":"running"},"truncated":false}},"sourceEventSeqs":[25],"surfaceOp":"append"}
 {"type":"step/end","seq":27,"time":1785730504688,"data":{"turn":1,"step":2}}
 {"type":"step/start","seq":28,"time":1785730504696,"data":{"turn":1,"step":3}}
 {"type":"assistant/chunk","seq":29,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}
@@ -33,9 +33,9 @@
 {"type":"assistant/chunk","seq":31,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"pty-read","name":"terminal_read","arguments":"{\"sessionId\":\"pty-1\",\"offset\":0,\"count\":20}"}}}}
 {"type":"assistant/chunk","seq":32,"time":1785498587473,"data":{"turn":1,"step":3,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
 {"type":"assistant/chunk","seq":33,"time":1785730504697,"data":{"turn":1,"step":3,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
-{"type":"assistant/message","seq":34,"time":1785730504697,"data":{"turn":1,"step":3,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-read","name":"terminal_read","arguments":"{\"sessionId\":\"pty-1\",\"offset\":0,\"count\":20}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"4e0d8eed-877d-4a5b-bd92-e7c4c8f2cf23"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[29,30,31,32,33],"surfaceOp":"append"}
+{"type":"assistant/message","seq":34,"time":1785730504697,"data":{"turn":1,"step":3,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-read","name":"terminal_read","arguments":"{\"sessionId\":\"pty-1\",\"offset\":0,\"count\":20}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"4138011c-7ea6-4a44-888d-e957ee7a234b"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[29,30,31,32,33],"surfaceOp":"append"}
 {"type":"tool/call","seq":35,"time":1785730504697,"data":{"turn":1,"step":3,"callId":"pty-read","name":"terminal_read","arguments":"{\"sessionId\":\"pty-1\",\"offset\":0,\"count\":20}"}}
-{"type":"tool/result","seq":36,"time":1785730504704,"data":{"turn":1,"step":3,"message":{"source":{"kind":"tool","callId":"pty-read"},"content":[{"type":"tool-result","toolCallId":"pty-read","content":[{"type":"text","text":"dsh> printf 'PTY_OK\\n'\nPTY_OK\ndsh> \n[lines: 0-3 of 3]"}],"isError":false}],"role":"user","id":"72be20be-6022-4363-b9bb-1d9f2cce0e20"}},"sourceEventSeqs":[35],"surfaceOp":"append"}
+{"type":"tool/result","seq":36,"time":1785730504704,"data":{"turn":1,"step":3,"message":{"source":{"kind":"tool","callId":"pty-read"},"content":[{"type":"tool-result","toolCallId":"pty-read","content":[{"type":"text","text":"dsh> printf 'PTY_OK\\n'\nPTY_OK\ndsh> \n[lines: 0-3 of 3]"}],"isError":false}],"role":"user","id":"f992ca67-207d-4c9b-b6ee-f83019ec76c5"}},"sourceEventSeqs":[35],"surfaceOp":"append"}
 {"type":"step/end","seq":37,"time":1785730504704,"data":{"turn":1,"step":3}}
 {"type":"step/start","seq":38,"time":1785730504712,"data":{"turn":1,"step":4}}
 {"type":"assistant/chunk","seq":39,"time":0,"data":{"turn":1,"step":4,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}
@@ -43,9 +43,9 @@
 {"type":"assistant/chunk","seq":41,"time":0,"data":{"turn":1,"step":4,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"pty-signal","name":"terminal_signal","arguments":"{\"sessionId\":\"pty-missing\",\"signal\":\"SIGINT\"}"}}}}
 {"type":"assistant/chunk","seq":42,"time":1785498587489,"data":{"turn":1,"step":4,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
 {"type":"assistant/chunk","seq":43,"time":1785730504713,"data":{"turn":1,"step":4,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
-{"type":"assistant/message","seq":44,"time":1785730504713,"data":{"turn":1,"step":4,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-signal","name":"terminal_signal","arguments":"{\"sessionId\":\"pty-missing\",\"signal\":\"SIGINT\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"87c2857f-5da3-4050-b081-e044e207be88"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[39,40,41,42,43],"surfaceOp":"append"}
+{"type":"assistant/message","seq":44,"time":1785730504713,"data":{"turn":1,"step":4,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-signal","name":"terminal_signal","arguments":"{\"sessionId\":\"pty-missing\",\"signal\":\"SIGINT\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"7439e01e-41de-4a39-a48e-690a3984a789"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[39,40,41,42,43],"surfaceOp":"append"}
 {"type":"tool/call","seq":45,"time":1785730504713,"data":{"turn":1,"step":4,"callId":"pty-signal","name":"terminal_signal","arguments":"{\"sessionId\":\"pty-missing\",\"signal\":\"SIGINT\"}"}}
-{"type":"tool/result","seq":46,"time":1785730504721,"data":{"turn":1,"step":4,"message":{"source":{"kind":"tool","callId":"pty-signal"},"content":[{"type":"tool-result","toolCallId":"pty-signal","content":[{"type":"text","text":"Error: unknown PTY session pty-missing"}],"isError":true}],"role":"user","id":"c6e3ff43-809c-4bd8-ba88-da7294c3385a"}},"sourceEventSeqs":[45],"surfaceOp":"append"}
+{"type":"tool/result","seq":46,"time":1785730504721,"data":{"turn":1,"step":4,"message":{"source":{"kind":"tool","callId":"pty-signal"},"content":[{"type":"tool-result","toolCallId":"pty-signal","content":[{"type":"text","text":"Error: unknown PTY session pty-missing"}],"isError":true}],"role":"user","id":"60ecf9fc-8420-45fd-8963-e8f903ab9242"}},"sourceEventSeqs":[45],"surfaceOp":"append"}
 {"type":"step/end","seq":47,"time":1785730504721,"data":{"turn":1,"step":4}}
 {"type":"step/start","seq":48,"time":1785730504730,"data":{"turn":1,"step":5}}
 {"type":"assistant/chunk","seq":49,"time":0,"data":{"turn":1,"step":5,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}
@@ -53,9 +53,9 @@
 {"type":"assistant/chunk","seq":51,"time":0,"data":{"turn":1,"step":5,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"pty-kill","name":"terminal_close","arguments":"{\"sessionId\":\"pty-1\"}"}}}}
 {"type":"assistant/chunk","seq":52,"time":1785498587503,"data":{"turn":1,"step":5,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
 {"type":"assistant/chunk","seq":53,"time":1785730504731,"data":{"turn":1,"step":5,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
-{"type":"assistant/message","seq":54,"time":1785730504731,"data":{"turn":1,"step":5,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-kill","name":"terminal_close","arguments":"{\"sessionId\":\"pty-1\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"5b53e0d5-1c68-4988-9fab-d885a9122fe8"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[49,50,51,52,53],"surfaceOp":"append"}
+{"type":"assistant/message","seq":54,"time":1785730504731,"data":{"turn":1,"step":5,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-kill","name":"terminal_close","arguments":"{\"sessionId\":\"pty-1\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"1ba884e6-5153-4618-934c-faab52475292"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[49,50,51,52,53],"surfaceOp":"append"}
 {"type":"tool/call","seq":55,"time":1785730504731,"data":{"turn":1,"step":5,"callId":"pty-kill","name":"terminal_close","arguments":"{\"sessionId\":\"pty-1\"}"}}
-{"type":"tool/result","seq":56,"time":1785730504738,"data":{"turn":1,"step":5,"message":{"source":{"kind":"tool","callId":"pty-kill"},"content":[{"type":"tool-result","toolCallId":"pty-kill","content":[{"type":"text","text":"closed terminal session pty-1"}],"isError":false}],"role":"user","id":"20e3fe4d-3d9d-4771-b75d-1f287fc20048"}},"sourceEventSeqs":[55],"surfaceOp":"append"}
+{"type":"tool/result","seq":56,"time":1785730504738,"data":{"turn":1,"step":5,"message":{"source":{"kind":"tool","callId":"pty-kill"},"content":[{"type":"tool-result","toolCallId":"pty-kill","content":[{"type":"text","text":"closed terminal session pty-1"}],"isError":false}],"role":"user","id":"c7e6e12b-49c1-4830-bea7-fa4fb4dd78bf"}},"sourceEventSeqs":[55],"surfaceOp":"append"}
 {"type":"step/end","seq":57,"time":1785730504738,"data":{"turn":1,"step":5}}
 {"type":"step/start","seq":58,"time":1785730504746,"data":{"turn":1,"step":6}}
 {"type":"assistant/chunk","seq":59,"time":0,"data":{"turn":1,"step":6,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}
@@ -63,9 +63,9 @@
 {"type":"assistant/chunk","seq":61,"time":0,"data":{"turn":1,"step":6,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"pty-list","name":"terminal_list","arguments":"{}"}}}}
 {"type":"assistant/chunk","seq":62,"time":1785498587517,"data":{"turn":1,"step":6,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
 {"type":"assistant/chunk","seq":63,"time":1785730504747,"data":{"turn":1,"step":6,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
-{"type":"assistant/message","seq":64,"time":1785730504747,"data":{"turn":1,"step":6,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-list","name":"terminal_list","arguments":"{}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"178fb9bc-859c-484d-8376-096a705de30a"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[59,60,61,62,63],"surfaceOp":"append"}
+{"type":"assistant/message","seq":64,"time":1785730504747,"data":{"turn":1,"step":6,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-list","name":"terminal_list","arguments":"{}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"73c85a86-45f3-4231-b4b0-6e28de66d93e"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[59,60,61,62,63],"surfaceOp":"append"}
 {"type":"tool/call","seq":65,"time":1785730504747,"data":{"turn":1,"step":6,"callId":"pty-list","name":"terminal_list","arguments":"{}"}}
-{"type":"tool/result","seq":66,"time":1785730504755,"data":{"turn":1,"step":6,"message":{"source":{"kind":"tool","callId":"pty-list"},"content":[{"type":"tool-result","toolCallId":"pty-list","content":[{"type":"text","text":"(no terminal sessions)"}],"isError":false}],"role":"user","id":"584673b3-9221-4a42-b9e4-69ce1b9f4d60"}},"sourceEventSeqs":[65],"surfaceOp":"append"}
+{"type":"tool/result","seq":66,"time":1785730504755,"data":{"turn":1,"step":6,"message":{"source":{"kind":"tool","callId":"pty-list"},"content":[{"type":"tool-result","toolCallId":"pty-list","content":[{"type":"text","text":"(no terminal sessions)"}],"isError":false}],"role":"user","id":"06c55e1a-c207-4895-8e52-a2c2ad3a980f"}},"sourceEventSeqs":[65],"surfaceOp":"append"}
 {"type":"step/end","seq":67,"time":1785730504755,"data":{"turn":1,"step":6}}
 {"type":"step/start","seq":68,"time":1785730504763,"data":{"turn":1,"step":7}}
 {"type":"assistant/chunk","seq":69,"time":0,"data":{"turn":1,"step":7,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
@@ -73,6 +73,6 @@
 {"type":"assistant/chunk","seq":71,"time":0,"data":{"turn":1,"step":7,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"DONE"}}}}
 {"type":"assistant/chunk","seq":72,"time":1785498587531,"data":{"turn":1,"step":7,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":3}}}}
 {"type":"assistant/chunk","seq":73,"time":1785730504764,"data":{"turn":1,"step":7,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}
-{"type":"assistant/message","seq":74,"time":1785730504764,"data":{"turn":1,"step":7,"message":{"role":"assistant","content":[{"type":"text","text":"DONE"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"23d457ba-b690-4bb9-b434-86f43c9f4da5"},"usage":{"inputTokens":10,"outputTokens":3}},"sourceEventSeqs":[69,70,71,72,73],"surfaceOp":"append"}
+{"type":"assistant/message","seq":74,"time":1785730504764,"data":{"turn":1,"step":7,"message":{"role":"assistant","content":[{"type":"text","text":"DONE"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"dffd880d-125b-45c3-b7cc-016159f5c043"},"usage":{"inputTokens":10,"outputTokens":3}},"sourceEventSeqs":[69,70,71,72,73],"surfaceOp":"append"}
 {"type":"step/end","seq":75,"time":1785730504764,"data":{"turn":1,"step":7}}
 {"type":"turn/end","seq":76,"time":1785730504764,"data":{"turn":1,"reason":{"kind":"completed"}}}
diff --git a/examples/headless-agent/tests/snapshots/subagent-settlement/child.expected.jsonl b/examples/headless-agent/tests/snapshots/subagent-settlement/child.expected.jsonl
new file mode 100644
index 0000000000..9d63ab152c
--- /dev/null
+++ b/examples/headless-agent/tests/snapshots/subagent-settlement/child.expected.jsonl
@@ -0,0 +1,20 @@
+{"type":"session","version":0,"id":"{{sessionId}}","createdAt":0,"cwd":"{{cwd}}","parentSession":"{{sessionId}}","origin":"subagent","delegationDepth":1}
+{"type":"subagent/descriptor","seq":0,"time":0,"data":{"version":2,"mode":"continuable","provider":"spawn","label":"Return child result","agentProvider":"deepseek-official","agentModel":"deepseek-v4-flash"}}
+{"type":"session/end-seed","seq":1,"time":0,"data":{}}
+{"type":"agent/inbox/spliced","seq":2,"time":0,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Reply with exactly CHILD_RESULT and nothing else. Do not call report."}],"source":{"kind":"user"},"role":"user","id":"{{sessionId}}"}]}}
+{"type":"turn/start","seq":3,"time":0,"data":{"turn":1}}
+{"type":"agent/inbox/spliced","seq":4,"time":0,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}
+{"type":"step/start","seq":5,"time":0,"data":{"turn":1,"step":1}}
+{"type":"user/message","seq":6,"time":0,"data":{"content":[{"type":"text","text":"Reply with exactly CHILD_RESULT and nothing else. Do not call report."}],"source":{"kind":"user"},"role":"user","id":"{{sessionId}}"},"surfaceOp":"append"}
+{"type":"user/message","seq":7,"time":0,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"{{sessionId}}"},"surfaceOp":"append"}
+{"type":"session/title","seq":8,"time":0,"data":{"title":"Reply with exactly CHILD_RESULT and","messageSeqs":[6],"source":{"kind":"fallback"}}}
+{"type":"request/header","seq":9,"time":0,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}}
+{"type":"request/context","seq":10,"time":0,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}}
+{"type":"assistant/chunk","seq":11,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
+{"type":"assistant/chunk","seq":12,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"text-delta","index":0,"text":"CHILD_RESULT"}}}
+{"type":"assistant/chunk","seq":13,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"CHILD_RESULT"}}}}
+{"type":"assistant/chunk","seq":14,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
+{"type":"assistant/chunk","seq":15,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}
+{"type":"assistant/message","seq":16,"time":0,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"CHILD_RESULT"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"{{sessionId}}"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[11,12,13,14,15],"surfaceOp":"append"}
+{"type":"step/end","seq":17,"time":0,"data":{"turn":1,"step":1}}
+{"type":"turn/end","seq":18,"time":0,"data":{"turn":1,"reason":{"kind":"completed"}}}
diff --git a/examples/headless-agent/tests/snapshots/subagent-settlement/child.replay.jsonl b/examples/headless-agent/tests/snapshots/subagent-settlement/child.replay.jsonl
new file mode 100644
index 0000000000..0f7362448a
--- /dev/null
+++ b/examples/headless-agent/tests/snapshots/subagent-settlement/child.replay.jsonl
@@ -0,0 +1,6 @@
+{"type":"session","version":0,"id":"subagent-settlement-child","createdAt":2,"delegationDepth":1}
+{"type":"assistant/chunk","seq":0,"time":1,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
+{"type":"assistant/chunk","seq":1,"time":2,"data":{"turn":1,"step":1,"chunk":{"type":"text-delta","index":0,"text":"CHILD_RESULT"}}}
+{"type":"assistant/chunk","seq":2,"time":3,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"CHILD_RESULT"}}}}
+{"type":"assistant/chunk","seq":3,"time":4,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}
+{"type":"assistant/chunk","seq":4,"time":5,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}
diff --git a/examples/headless-agent/tests/snapshots/subagent-settlement/parent.override.json b/examples/headless-agent/tests/snapshots/subagent-settlement/parent.override.json
new file mode 100644
index 0000000000..bdb77f505e
--- /dev/null
+++ b/examples/headless-agent/tests/snapshots/subagent-settlement/parent.override.json
@@ -0,0 +1,32 @@
+[
+  {
+    "kind": "chunks",
+    "chunks": [
+      { "type": "block-start", "index": 0, "blockType": "tool-call" },
+      { "type": "tool-call-delta", "index": 0, "id": "start-child", "name": "subagent", "argumentsDelta": "{\"description\":\"Return child result\",\"prompt\":\"Reply with exactly CHILD_RESULT and nothing else. Do not call report.\",\"run_in_background\":true}" },
+      { "type": "block-end", "index": 0, "block": { "type": "tool-call", "id": "start-child", "name": "subagent", "arguments": "{\"description\":\"Return child result\",\"prompt\":\"Reply with exactly CHILD_RESULT and nothing else. Do not call report.\",\"run_in_background\":true}" } },
+      { "type": "usage", "usage": { "inputTokens": 10, "outputTokens": 5 } },
+      { "type": "finish", "reason": { "kind": "tool-calls" } }
+    ]
+  },
+  {
+    "kind": "chunks",
+    "chunks": [
+      { "type": "block-start", "index": 0, "blockType": "text" },
+      { "type": "text-delta", "index": 0, "text": "STARTED" },
+      { "type": "block-end", "index": 0, "block": { "type": "text", "text": "STARTED" } },
+      { "type": "usage", "usage": { "inputTokens": 10, "outputTokens": 5 } },
+      { "type": "finish", "reason": { "kind": "stop" } }
+    ]
+  },
+  {
+    "kind": "chunks",
+    "chunks": [
+      { "type": "block-start", "index": 0, "blockType": "text" },
+      { "type": "text-delta", "index": 0, "text": "PARENT_RECEIVED_CHILD_RESULT" },
+      { "type": "block-end", "index": 0, "block": { "type": "text", "text": "PARENT_RECEIVED_CHILD_RESULT" } },
+      { "type": "usage", "usage": { "inputTokens": 10, "outputTokens": 5 } },
+      { "type": "finish", "reason": { "kind": "stop" } }
+    ]
+  }
+]
diff --git a/examples/headless-agent/tests/snapshots/subagent-settlement/parent.replay.jsonl b/examples/headless-agent/tests/snapshots/subagent-settlement/parent.replay.jsonl
new file mode 100644
index 0000000000..b4216aa0ed
--- /dev/null
+++ b/examples/headless-agent/tests/snapshots/subagent-settlement/parent.replay.jsonl
@@ -0,0 +1 @@
+{"type":"session","version":0,"id":"subagent-settlement-parent","createdAt":1,"delegationDepth":0}
diff --git a/examples/headless-agent/tests/snapshots/subagent-settlement/stream-json.expected.jsonl b/examples/headless-agent/tests/snapshots/subagent-settlement/stream-json.expected.jsonl
new file mode 100644
index 0000000000..bc181b78ad
--- /dev/null
+++ b/examples/headless-agent/tests/snapshots/subagent-settlement/stream-json.expected.jsonl
@@ -0,0 +1,38 @@
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"agent/inbox/spliced","seq":0,"time":0,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Start one continuable background subagent and answer from its completion notice. Do not call list_agents, send_message, task_output, or task_list."}],"source":{"kind":"user"},"role":"user","id":"{{sessionId}}"}]}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"turn/start","seq":1,"time":0,"data":{"turn":1}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"agent/inbox/spliced","seq":2,"time":0,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"step/start","seq":3,"time":0,"data":{"turn":1,"step":1}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"user/message","seq":4,"time":0,"data":{"content":[{"type":"text","text":"Start one continuable background subagent and answer from its completion notice. Do not call list_agents, send_message, task_output, or task_list."}],"source":{"kind":"user"},"role":"user","id":"{{sessionId}}"},"surfaceOp":"append"}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"session/title","seq":5,"time":0,"data":{"title":"Start one continuable background subagen","messageSeqs":[4],"source":{"kind":"fallback"}}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"request/header","seq":6,"time":0,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"request/context","seq":7,"time":0,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":8,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":9,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"tool-call-delta","index":0,"id":"start-child","name":"subagent","argumentsDelta":"{\"description\":\"Return child result\",\"prompt\":\"Reply with exactly CHILD_RESULT and nothing else. Do not call report.\",\"run_in_background\":true}"}}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":10,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"start-child","name":"subagent","arguments":"{\"description\":\"Return child result\",\"prompt\":\"Reply with exactly CHILD_RESULT and nothing else. Do not call report.\",\"run_in_background\":true}"}}}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":11,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":12,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/message","seq":13,"time":0,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"start-child","name":"subagent","arguments":"{\"description\":\"Return child result\",\"prompt\":\"Reply with exactly CHILD_RESULT and nothing else. Do not call report.\",\"run_in_background\":true}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"{{sessionId}}"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[8,9,10,11,12],"surfaceOp":"append"}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"tool/call","seq":14,"time":0,"data":{"turn":1,"step":1,"callId":"start-child","name":"subagent","arguments":"{\"description\":\"Return child result\",\"prompt\":\"Reply with exactly CHILD_RESULT and nothing else. Do not call report.\",\"run_in_background\":true}"}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"tool/result","seq":15,"time":0,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"start-child"},"content":[{"type":"tool-result","toolCallId":"start-child","content":[{"type":"text","text":"started subagent {{sessionId}}"}],"isError":false}],"role":"user","id":"{{sessionId}}"}},"sourceEventSeqs":[14],"surfaceOp":"append"}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"step/end","seq":16,"time":0,"data":{"turn":1,"step":1}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"agent/inbox/spliced","seq":17,"time":0,"data":{"target":"next-step","start":0,"inserted":[{"content":[{"type":"text","text":"Background subagent {{sessionId}} finished and will do no further work unless you send it more."},{"type":"text","text":"Its closing message:"},{"type":"text","text":"CHILD_RESULT"}],"source":{"kind":"subagent-settled","form":"notice","summary":"Background subagent {{sessionId}} finished and will do no further work unless you send it more.","senderSessionId":"{{sessionId}}"},"role":"user","id":"{{sessionId}}"}]}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"step/start","seq":18,"time":0,"data":{"turn":1,"step":2}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":19,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":20,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"text-delta","index":0,"text":"STARTED"}}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":21,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"STARTED"}}}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":22,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":23,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/message","seq":24,"time":0,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"text","text":"STARTED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"{{sessionId}}"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"step/end","seq":25,"time":0,"data":{"turn":1,"step":2}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"agent/inbox/spliced","seq":26,"time":0,"data":{"target":"next-step","start":0,"removedCount":1,"inserted":[]}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"step/start","seq":27,"time":0,"data":{"turn":1,"step":3}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"user/message","seq":28,"time":0,"data":{"content":[{"type":"text","text":"Background subagent {{sessionId}} finished and will do no further work unless you send it more."},{"type":"text","text":"Its closing message:"},{"type":"text","text":"CHILD_RESULT"}],"source":{"kind":"subagent-settled","form":"notice","summary":"Background subagent {{sessionId}} finished and will do no further work unless you send it more.","senderSessionId":"{{sessionId}}"},"role":"user","id":"{{sessionId}}"},"surfaceOp":"append"}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":29,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":30,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"text-delta","index":0,"text":"PARENT_RECEIVED_CHILD_RESULT"}}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":31,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"PARENT_RECEIVED_CHILD_RESULT"}}}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":32,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":33,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/message","seq":34,"time":0,"data":{"turn":1,"step":3,"message":{"role":"assistant","content":[{"type":"text","text":"PARENT_RECEIVED_CHILD_RESULT"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"{{sessionId}}"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[29,30,31,32,33],"surfaceOp":"append"}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"step/end","seq":35,"time":0,"data":{"turn":1,"step":3}}}
+{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"turn/end","seq":36,"time":0,"data":{"turn":1,"reason":{"kind":"completed"}}}}
+{"type":"result","sessionId":"{{sessionId}}","output":"PARENT_RECEIVED_CHILD_RESULT","usage":{"inputTokens":30,"outputTokens":15}}
diff --git a/knip.json b/knip.json
index e418c8ce3d..e10b466941 100644
--- a/knip.json
+++ b/knip.json
@@ -38,6 +38,7 @@
         "headless-agent/tests/fixtures/semantic-checkpoint-agent.ts",
         "headless-agent/tests/fixtures/subagent-diagnostic-agent.ts",
         "headless-agent/tests/fixtures/subagent-inheritance-agent.ts",
+        "headless-agent/tests/fixtures/subagent-settlement-fence.ts",
         "headless-agent/tests/fixtures/workspace-context-resume-agent.ts",
         "headless-agent/tests/fixtures/goal-domain/seed-goal.ts",
         "headless-agent/tests/fixtures/time-context-driver.ts",
@@ -50,6 +51,7 @@
         "acp-agent/tests/fixtures/parent-sandbox-override.ts",
         "acp-agent/tests/fixtures/partial-landlock-sandbox.ts",
         "acp-agent/tests/fixtures/subagent-durability-failure.ts",
+        "acp-agent/tests/fixtures/subagent-report-fence.ts",
         "acp-agent/tests/fixtures/subagent-settlement-marker.ts",
         "acp-agent/tests/fixtures/workspace-context-compaction.ts",
         "acp-agent/tests/fixtures/subagent/subagent-acp/mock-delegating-llm.ts",
diff --git a/packages/bundle/base/cordis.patch.yml b/packages/bundle/base/cordis.patch.yml
index 2fb5ff10d3..b9448ae6f2 100644
--- a/packages/bundle/base/cordis.patch.yml
+++ b/packages/bundle/base/cordis.patch.yml
@@ -311,12 +311,16 @@
         toolName: subagent
         backgroundMode: continuable
 
+    # Fork stays one-shot: a continuable child's `report` tool and prompt
+    # section precede the inherited history a fork exists to reuse; one-shot
+    # fork children install neither, keeping the parent's request prefix.
+    # See .agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md.
     - id: tool-subagent-fork
       name: '@deepseek-ai/dsh-tool-subagent'
       config:
         provider: fork
         toolName: subagent_fork
-        backgroundMode: continuable
+        backgroundMode: one-shot
 
     # Optional direct-child return channel; absent from roots and one-shot agents.
     - id: tool-subagent-report
diff --git a/packages/core/agent/README.i18n.yaml b/packages/core/agent/README.i18n.yaml
index 477c3a5cb6..8022432bc1 100644
--- a/packages/core/agent/README.i18n.yaml
+++ b/packages/core/agent/README.i18n.yaml
@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write packages/core/agent/README.md
-README.md: 0fb65b94a3b311aa9f0df09d39dd937cba4cc7b4
-README.zh.md: 44f67483343a98c280317793ece544bd0b984596
+README.md: 61294df5cca1b03f9f158678f2992a6e4fbaaffd
+README.zh.md: 5576a07330926cfd8fd5bd42c8a0351a375f035c
diff --git a/packages/core/agent/README.md b/packages/core/agent/README.md
index 0fb65b94a3..61294df5cc 100644
--- a/packages/core/agent/README.md
+++ b/packages/core/agent/README.md
@@ -58,6 +58,8 @@ Inbox live notifications are deliberately per-message and minimal: `agent/inbox/
 
 Turn and step boundaries and the model token stream are durable `session/event` facts rather than mirrored `agent/*` notifications. Consumers read `turn/*`, `step/*`, and `assistant/chunk` from the session feed; tool policy and outcome observation belong to the complete pipeline documented by [`dsh-tools`](../tools/README.md).
 
+`foldConsumedWork(events)` reads that feed back for the one question the turn sequence cannot answer alone: what became of the work a log consumed. It returns the latest `turn/end` that accounts for consumed work — a turn that entered a model step, or one that claimed inbox input and then failed, was stopped, or was rejected before reaching one — plus whether accepted work was later cancelled out of the inbox unrun. Both facts come from the log, so a cancellation reads the same whichever owner issued it. A no-step turn that took nothing, or emptied its claim and completed, describes no work and is skipped; a `blocked` end over claimed input is an account, because rejection discarded that input.
+
 ### Agent interface (`types.ts`)
 
 The handle every plugin programs against:
diff --git a/packages/core/agent/README.zh.md b/packages/core/agent/README.zh.md
index 44f6748334..5576a07330 100644
--- a/packages/core/agent/README.zh.md
+++ b/packages/core/agent/README.zh.md
@@ -58,6 +58,8 @@ inbox 的实时通知刻意采用逐消息的最小载荷：`agent/inbox/inserte
 
 轮次和步骤边界以及模型 token 流是持久 `session/event` 事实，而不是镜像的 `agent/*` 通知。消费方从会话事件流读取 `turn/*`、`step/*` 和 `assistant/chunk`；工具策略与结果观测属于 [`dsh-tools`](../tools/README.md) 记录的完整流水线。
 
+`foldConsumedWork(events)` 把这条事件流读回来，回答仅凭轮次序列无法回答的那个问题：一份日志消费掉的工作最终怎样了。它返回能够为已消费工作作出交代的最新 `turn/end`——即进入过模型 step 的轮次，或者认领了 inbox 输入、但在进入 step 之前失败、被停下或被拒绝的轮次——并额外给出「已接受的工作此后是否被从 inbox 中取消且从未运行」。两项事实都来自日志，因此无论由哪个所有者发起取消，读出来都一样。没有取走任何输入、或认领批次被改写清空后正常结束的无 step 轮次不描述工作，会被跳过；认领过输入、以 `blocked` 结束的轮次则是一份交代，因为拒绝把这些输入一并丢弃了。
+
 ### Agent 接口（`types.ts`）
 
 每个插件面向的 handle：
diff --git a/packages/core/agent/src/consumed-work.ts b/packages/core/agent/src/consumed-work.ts
new file mode 100644
index 0000000000..b86ac836ef
--- /dev/null
+++ b/packages/core/agent/src/consumed-work.ts
@@ -0,0 +1,108 @@
+/**
+ * How one agent log accounts for the work it consumed.
+ *
+ * The turn and step vocabulary alone cannot answer this. A turn that stops
+ * before its first step leaves a `turn/end` shaped exactly like the balanced
+ * no-op turns a rejection or an empty claim produces, so reading turns in
+ * isolation either credits cut-short work as finished or convicts every no-op.
+ * The missing fact is the inbox's own record: {@link Inbox} logs each mutation
+ * with `removedCount` and marks a cancellation `outcome: 'canceled'`, which
+ * separates a turn claiming its input from work being dropped unrun.
+ *
+ * @module @deepseek-ai/dsh-agent/consumed-work
+ */
+
+import type { SessionEvent, TurnEndReason } from '@deepseek-ai/dsh-session'
+
+/** How one agent log accounts for the work it consumed. */
+export interface ConsumedWork {
+  /**
+   * The latest closed turn that accounts for consumed work: one that entered a
+   * model step, or one that claimed inbox input and then failed, was stopped,
+   * or was rejected. Absent when no turn closed over any work.
+   */
+  readonly end?: SessionEvent<'turn/end'>
+  /**
+   * Whether accepted work was cancelled out of the inbox, unrun, after that
+   * turn. This is the only account of input a cancellation took before any turn
+   * could open over it — no `turn/end` describes it.
+   */
+  readonly droppedUnrun: boolean
+}
+
+/**
+ * Whether a turn that consumed input but never reached a step ends in a way
+ * that accounts for that input. Only a `completed` end does not: it had
+ * nothing left to run once its claim was rewritten away. A `blocked` end is
+ * that input's ending too — the pre-step rejection that produced it discarded
+ * the claimed messages, so the work it took will never run.
+ * @param reason - the turn's recorded ending.
+ * @returns whether the ending accounts for the input the turn took.
+ */
+function accountsForClaim(reason: TurnEndReason): boolean {
+  switch (reason.kind) {
+    case 'completed':
+      return false
+    case 'blocked':
+    case 'aborted':
+    case 'interrupted':
+    case 'error':
+      return true
+    /* v8 ignore next 4 -- unreachable: the one unnamed built-in, `max-tokens`, requires a step,
+     * so its turn short-circuits as stepped before this call, and `TurnEndReasonMap` is
+     * merge-extensible, so a backend-added variant cannot be listed; an unnameable ending over
+     * consumed input must not read as success. */
+    default:
+      return true
+  }
+}
+
+/**
+ * Fold one agent log, or an owned suffix of one, into its account of consumed
+ * work. Single pass, and every input is the log itself: no caller has to sample
+ * live state before cancelling, so a cancellation issued by anyone — the owner's
+ * teardown, an ancestor's interrupt, an unloading plugin — reads the same.
+ * @param events - the log, or an owned suffix, to fold.
+ * @returns the accounting turn when one closed, and whether work was dropped unrun after it.
+ */
+export function foldConsumedWork(events: readonly SessionEvent[]): ConsumedWork {
+  const stepped = new Set<number>()
+  const claimed = new Set<number>()
+  let open: number | undefined
+  let end: SessionEvent<'turn/end'> | undefined
+  let droppedUnrun = false
+  for (const event of events) {
+    switch (event.type) {
+      case 'turn/start':
+        open = event.data.turn
+        break
+      case 'step/start':
+        stepped.add(event.data.turn)
+        break
+      case 'agent/inbox/spliced': {
+        const { removedCount, outcome, inserted } = event.data
+        if (removedCount === undefined) break
+        // A replacement keeps the work pending under a new identity, so only a
+        // cancellation that leaves nothing behind drops it.
+        if (outcome === 'canceled') droppedUnrun ||= inserted.length === 0
+        // Claims are the loop's own step-boundary reads, always inside a turn.
+        else if (open !== undefined) claimed.add(open)
+        break
+      }
+      case 'turn/end': {
+        const { turn, reason } = event.data
+        open = undefined
+        if (stepped.delete(turn) || (claimed.delete(turn) && accountsForClaim(reason))) {
+          end = event
+          // Anything dropped before this turn closed is what its own ending
+          // reports; only a later drop is still unaccounted for.
+          droppedUnrun = false
+        }
+        break
+      }
+      default:
+        break
+    }
+  }
+  return { ...end === undefined ? {} : { end }, droppedUnrun }
+}
diff --git a/packages/core/agent/src/index.ts b/packages/core/agent/src/index.ts
index dffc1f01b3..5211942f11 100644
--- a/packages/core/agent/src/index.ts
+++ b/packages/core/agent/src/index.ts
@@ -18,6 +18,7 @@ import type { Agent, AgentOptions } from './runtime-types.ts'
 export * from './runtime-types.ts'
 export * from './types.ts'
 export * from './inbox.ts'
+export * from './consumed-work.ts'
 export * from './model-selection.ts'
 export { agentCarrier, agentEvents, assembleContextFor, emitAgentEvent } from './dispatch.ts'
 export type { AgentEventDispatch, AgentSubjectEvent } from './dispatch.ts'
diff --git a/packages/core/agent/tests/consumed-work.spec.ts b/packages/core/agent/tests/consumed-work.spec.ts
new file mode 100644
index 0000000000..3b79d67620
--- /dev/null
+++ b/packages/core/agent/tests/consumed-work.spec.ts
@@ -0,0 +1,160 @@
+import { describe, expect, it } from 'vitest'
+import { createUserMessage } from '@deepseek-ai/dsh-llm'
+import { Session, SessionId } from '@deepseek-ai/dsh-session'
+import type { TurnEndReason } from '@deepseek-ai/dsh-session'
+import { foldConsumedWork } from '@deepseek-ai/dsh-agent'
+
+/** One pending message, as the inbox records it. */
+function message(text: string) {
+  return createUserMessage({ content: [{ type: 'text', text }], source: { kind: 'user' } })
+}
+
+/** Log an accepted message the way `Inbox.append()` does. */
+function accept(session: Session, text: string): void {
+  session.append('agent/inbox/spliced', { target: 'next-turn', start: 0, inserted: [message(text)] })
+}
+
+/** Log the step-boundary read of one pending message, as `Inbox.claim()` does. */
+function claim(session: Session): void {
+  session.append('agent/inbox/spliced', { target: 'next-turn', start: 0, removedCount: 1, inserted: [] })
+}
+
+/** Log a cancellation of one pending message, as `Inbox.clear()` does. */
+function cancelPending(session: Session): void {
+  session.append('agent/inbox/spliced', {
+    target: 'next-turn', start: 0, removedCount: 1, inserted: [], outcome: 'canceled',
+  })
+}
+
+/** Run one whole turn that reached a model step. */
+function steppedTurn(session: Session, turn: number, reason: TurnEndReason): void {
+  session.append('turn/start', { turn })
+  claim(session)
+  session.append('step/start', { turn, step: 1 })
+  session.append('step/end', { turn, step: 1 })
+  session.append('turn/end', { turn, reason })
+}
+
+describe('foldConsumedWork', () => {
+  it('reports nothing for a log that consumed no work', () => {
+    const session = Session.create(SessionId('empty'))
+    accept(session, 'queued')
+
+    expect(foldConsumedWork(session.events)).toEqual({ droppedUnrun: false })
+  })
+
+  it('reports the latest turn that entered a model step', () => {
+    const session = Session.create(SessionId('stepped'))
+    steppedTurn(session, 1, { kind: 'completed' })
+    steppedTurn(session, 2, { kind: 'max-tokens' })
+
+    expect(foldConsumedWork(session.events).end?.data)
+      .toEqual({ turn: 2, reason: { kind: 'max-tokens' } })
+  })
+
+  it('reports a turn that claimed its input and then failed before any step', () => {
+    const session = Session.create(SessionId('failed-claim'))
+    steppedTurn(session, 1, { kind: 'completed' })
+    // The step boundary runs the durability checkpoint and prompt assembly, so a
+    // turn can take its input and then fail without entering a step.
+    session.append('turn/start', { turn: 2 })
+    claim(session)
+    session.append('turn/end', { turn: 2, reason: { kind: 'error', error: { message: 'ENOSPC', code: 'UNKNOWN' } } })
+
+    expect(foldConsumedWork(session.events).end?.data.turn).toBe(2)
+  })
+
+  it('reports a turn that claimed its input and was then stopped before any step', () => {
+    const session = Session.create(SessionId('stopped-claim'))
+    steppedTurn(session, 1, { kind: 'completed' })
+    session.append('turn/start', { turn: 2 })
+    claim(session)
+    session.append('turn/end', { turn: 2, reason: { kind: 'aborted', reason: { kind: 'user' } } })
+
+    expect(foldConsumedWork(session.events).end?.data.turn).toBe(2)
+  })
+
+  it('ignores a turn stopped, failed, or rejected without taking any input', () => {
+    const session = Session.create(SessionId('no-claim'))
+    steppedTurn(session, 1, { kind: 'completed' })
+    session.append('turn/start', { turn: 2 })
+    session.append('turn/end', { turn: 2, reason: { kind: 'aborted', reason: { kind: 'parent' } } })
+    session.append('turn/start', { turn: 3 })
+    session.append('turn/end', { turn: 3, reason: { kind: 'error', error: { message: 'x', code: 'UNKNOWN' } } })
+    session.append('turn/start', { turn: 4 })
+    session.append('turn/end', { turn: 4, reason: { kind: 'blocked' } })
+
+    // None of these turns describes work: they opened, found nothing of their own, and closed.
+    expect(foldConsumedWork(session.events).end?.data.turn).toBe(1)
+  })
+
+  it('reports a turn whose claimed input a pre-step rejection discarded', () => {
+    const session = Session.create(SessionId('rejected-claim'))
+    steppedTurn(session, 1, { kind: 'completed' })
+    session.append('turn/start', { turn: 2 })
+    claim(session)
+    session.append('turn/end', { turn: 2, reason: { kind: 'blocked' } })
+
+    // Rejection does not retain the claimed messages, so the `blocked` end is
+    // the only account of input that will never run.
+    expect(foldConsumedWork(session.events).end?.data.turn).toBe(2)
+  })
+
+  it('ignores a claim its own turn emptied', () => {
+    const session = Session.create(SessionId('emptied-claim'))
+    steppedTurn(session, 1, { kind: 'completed' })
+    session.append('turn/start', { turn: 2 })
+    claim(session)
+    session.append('turn/end', { turn: 2, reason: { kind: 'completed' } })
+
+    // An emptied claim ran nothing and dropped nothing: a listener rewrote the
+    // batch away, which is not this log's account of the work.
+    expect(foldConsumedWork(session.events).end?.data.turn).toBe(1)
+  })
+
+  it('credits a claim with no open turn to no turn at all', () => {
+    const session = Session.create(SessionId('mid-turn-suffix'))
+    steppedTurn(session, 1, { kind: 'completed' })
+    // An owned suffix can begin inside a turn whose start it does not contain,
+    // so a claim may appear with no turn to attribute it to.
+    claim(session)
+    session.append('turn/end', { turn: 2, reason: { kind: 'aborted', reason: { kind: 'user' } } })
+
+    expect(foldConsumedWork(session.events).end?.data.turn).toBe(1)
+  })
+
+  it('reports work cancelled out of the inbox after the last accounting turn', () => {
+    const session = Session.create(SessionId('dropped'))
+    steppedTurn(session, 1, { kind: 'completed' })
+    accept(session, 'never runs')
+    cancelPending(session)
+
+    // No turn opened over it, so only the cancellation says the work was cut short.
+    expect(foldConsumedWork(session.events)).toEqual({
+      end: session.events.find(event => event.type === 'turn/end'),
+      droppedUnrun: true,
+    })
+  })
+
+  it('keeps a replacement pending rather than counting it as dropped', () => {
+    const session = Session.create(SessionId('replaced'))
+    steppedTurn(session, 1, { kind: 'completed' })
+    session.append('agent/inbox/spliced', {
+      target: 'next-turn', start: 0, removedCount: 1, inserted: [message('rewritten')], outcome: 'canceled',
+    })
+
+    expect(foldConsumedWork(session.events).droppedUnrun).toBe(false)
+  })
+
+  it('lets a later accounting turn absorb an earlier drop', () => {
+    const session = Session.create(SessionId('absorbed'))
+    steppedTurn(session, 1, { kind: 'completed' })
+    cancelPending(session)
+    steppedTurn(session, 2, { kind: 'completed' })
+
+    expect(foldConsumedWork(session.events)).toEqual({
+      end: session.events.findLast(event => event.type === 'turn/end'),
+      droppedUnrun: false,
+    })
+  })
+})
diff --git a/packages/core/session/src/index.ts b/packages/core/session/src/index.ts
index 5251ca9408..477d08ae19 100644
--- a/packages/core/session/src/index.ts
+++ b/packages/core/session/src/index.ts
@@ -34,27 +34,6 @@ export { deriveEventMessage, foldSurface, isAppendSurfaceEvent, isReplacementSur
 export { canonicalHeader, foldRequestHeader, headerEquals } from './request-header.ts'
 export { KNOWN_SESSION_EVENT_TYPES } from './known-event-types.ts'
 
-/**
- * Find the latest closed turn that entered at least one model step, ignoring
- * balanced no-step turns produced by rejection, empty input, or cancellation.
- * @param events - session events, or an owned suffix, to inspect.
- * @returns the latest matching turn end, or `undefined`.
- */
-export function findLastMessageTurnEnd(
-  events: readonly SessionEvent[],
-): SessionEvent<'turn/end'> | undefined {
-  const steppedTurns = new Set<number>()
-  let latest: SessionEvent<'turn/end'> | undefined
-  for (const event of events) {
-    if (event.type === 'step/start') {
-      steppedTurns.add(event.data.turn)
-      continue
-    }
-    if (event.type === 'turn/end' && steppedTurns.delete(event.data.turn)) latest = event
-  }
-  return latest
-}
-
 declare module '@deepseek-ai/cordis' {
   interface Context {
     sessions: SessionStore
diff --git a/packages/core/session/tests/session.spec.ts b/packages/core/session/tests/session.spec.ts
index 7fe7ee01c7..d29624c030 100644
--- a/packages/core/session/tests/session.spec.ts
+++ b/packages/core/session/tests/session.spec.ts
@@ -7,28 +7,11 @@ import SessionStore, {
   Session,
   SessionEvent,
   SessionId,
-  findLastMessageTurnEnd,
   snapshotSessionEvent,
 } from '@deepseek-ai/dsh-session'
 import type { CreateSessionOptions, SessionEventType, SessionHeader, SessionSurface, TodoItem } from '@deepseek-ai/dsh-session'
 
 describe('Session', () => {
-  it('finds the latest closed turn that entered a model step', () => {
-    const session = Session.create(SessionId('last-message-turn'))
-    session.append('turn/start', { turn: 1 })
-    session.append('turn/end', { turn: 1, reason: { kind: 'blocked' } })
-
-    expect(findLastMessageTurnEnd(session.events)).toBeUndefined()
-
-    session.append('turn/start', { turn: 2 })
-    session.append('step/start', { turn: 2, step: 1 })
-    session.append('step/end', { turn: 2, step: 1 })
-    session.append('turn/end', { turn: 2, reason: { kind: 'max-tokens' } })
-
-    expect(findLastMessageTurnEnd(session.events)?.data)
-      .toEqual({ turn: 2, reason: { kind: 'max-tokens' } })
-  })
-
   it('exposes one stable readonly surface view', () => {
     const session = Session.create(SessionId('surface-view'))
     const surface = session.surface
diff --git a/packages/subagent/subagent-fork/README.i18n.yaml b/packages/subagent/subagent-fork/README.i18n.yaml
index e40cf9901f..ca23eabb84 100644
--- a/packages/subagent/subagent-fork/README.i18n.yaml
+++ b/packages/subagent/subagent-fork/README.i18n.yaml
@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write packages/subagent/subagent-fork/README.md
-README.md: 55475aee7841e91960de79887dfe9bf37afdf9da
-README.zh.md: 379218970c406162309659277a4fcb2da885d9ea
+README.md: 2bd72058ab7d112f8f317a33842fc4d95b719017
+README.zh.md: 40f9e34c5a8ae8c74de2ae0c9e4676866f0bd083
diff --git a/packages/subagent/subagent-fork/README.md b/packages/subagent/subagent-fork/README.md
index 55475aee78..2bd72058ab 100644
--- a/packages/subagent/subagent-fork/README.md
+++ b/packages/subagent/subagent-fork/README.md
@@ -39,7 +39,7 @@ Forking duplicates retained completed history into separate child requests; the
 
 #### KV Cache effect
 
-The child may reuse the inherited byte-identical prefix under the same provider and model. Persona, tool-filter, generated-SDK, or route changes may invalidate reuse before inherited history; later child history is append-only.
+The child may reuse the inherited byte-identical prefix under the same provider and model. Persona, tool-filter, generated-SDK, or route changes may invalidate reuse before inherited history; later child history is append-only. Shipped compositions therefore bind this provider to `backgroundMode: one-shot`, because a continuable child additionally carries the child-scoped `report` tool and its prompt section — deltas that precede the inherited history and so invalidate all of it ([the fork-one-shot Agent Note](../../../.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md)).
 
 ### Parent tool result, indirectly
 
@@ -58,3 +58,4 @@ Append-only; newly visible content follows the reusable request prefix and does
 ## Known Limitations and Deferred Work
 
 - **The seed is a one-time snapshot** — the child sees the parent's completed turns as of the fork and nothing the parent logs afterwards; there is no live context sharing.
+- **No shipped composition creates a continuable fork child** — `prepareContinuable` remains implemented and the seam accepts it, but every shipped `cordis.yml` sets `backgroundMode: one-shot` on the fork delegation tool, so the provider's continuable path has no production caller. Reopening it requires the child's system prompt and tool schemas to match the parent's byte for byte, which the [`report` return channel](../tool-subagent-report/README.md) currently prevents. Rationale and the reintroduction condition: [the fork-one-shot Agent Note](../../../.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md).
diff --git a/packages/subagent/subagent-fork/README.zh.md b/packages/subagent/subagent-fork/README.zh.md
index 379218970c..40f9e34c5a 100644
--- a/packages/subagent/subagent-fork/README.zh.md
+++ b/packages/subagent/subagent-fork/README.zh.md
@@ -39,7 +39,7 @@ fork 会把保留的已完成历史复制到独立的子 agent 请求中；随
 
 #### KV Cache 影响
 
-在提供方和模型相同的前提下，子 agent 可以复用继承的逐字节相同前缀。persona、工具过滤、生成 SDK 或路由变化可能在继承历史之前使复用失效；后续子 agent 历史仅追加。
+在提供方和模型相同的前提下，子 agent 可以复用继承的逐字节相同前缀。persona、工具过滤、生成 SDK 或路由变化可能在继承历史之前使复用失效；后续子 agent 历史仅追加。因此随附组合把本提供方绑定为 `backgroundMode: one-shot`：可继续子 agent 还会额外携带作用域局部的 `report` 工具及其提示词 section，而这些增量位于继承历史之前，会使继承历史整体失效（见 [fork 保持 one-shot 的 Agent Note](../../../.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md)）。
 
 ### 父 agent 工具结果（间接）
 
@@ -58,3 +58,4 @@ fork 会把保留的已完成历史复制到独立的子 agent 请求中；随
 ## 已知限制与暂缓事项
 
 - **初始内容是一次性快照**：子 agent 只能看到 fork 时父 agent 已完成的轮次，看不到父 agent 此后记录的任何内容；不会实时共享上下文。
+- **没有任何随附组合会创建可继续的 fork 子 agent**：`prepareContinuable` 仍然实现完好，seam 也接受它，但每份随附的 `cordis.yml` 都在 fork 委派工具上设置 `backgroundMode: one-shot`，因此该提供方的可继续路径没有生产调用方。重新开放它需要子 agent 的系统提示词与工具 schema 与父 agent 逐字节一致，而这一点目前被 [`report` 返回通道](../tool-subagent-report/README.md)阻止。理由与重新开放条件见 [fork 保持 one-shot 的 Agent Note](../../../.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md)。
diff --git a/packages/subagent/subagent-fork/src/index.ts b/packages/subagent/subagent-fork/src/index.ts
index 0a3ea21aa2..b4d3d51144 100644
--- a/packages/subagent/subagent-fork/src/index.ts
+++ b/packages/subagent/subagent-fork/src/index.ts
@@ -74,6 +74,12 @@ class ForkProvider implements SubagentProvider {
     })
   }
 
+  // TODO(fork-continuable-prefix-reuse): no shipped composition calls this —
+  // they bind fork to `backgroundMode: one-shot` because a continuable child's
+  // `report` tool and prompt section precede the inherited history, defeating
+  // the prefix reuse a fork exists for. Reopening needs a byte-identical child
+  // system prompt and tool schemas; see issue #2124 and
+  // .agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md.
   prepareContinuable(request: ContinuableCreateRequest): Promise<ContinuableCreateSpec> {
     // The fork prefix is captured ONCE, at creation: it becomes part of the
     // child's own durable transcript, so a later cold resume replays that
diff --git a/packages/subagent/subagent-inprocess/src/index.ts b/packages/subagent/subagent-inprocess/src/index.ts
index b0ab7425a5..0e7942fa55 100644
--- a/packages/subagent/subagent-inprocess/src/index.ts
+++ b/packages/subagent/subagent-inprocess/src/index.ts
@@ -13,8 +13,9 @@
 
 import { randomUUID } from 'node:crypto'
 import type { Context } from '@deepseek-ai/cordis'
+import { foldConsumedWork } from '@deepseek-ai/dsh-agent'
 import type { Agent, AgentHandle } from '@deepseek-ai/dsh-agent'
-import { findLastMessageTurnEnd, SessionId, type SessionEvent, type TurnEndReason } from '@deepseek-ai/dsh-session'
+import { SessionId, type SessionEvent, type TurnEndReason } from '@deepseek-ai/dsh-session'
 import { createUserMessage, type ContentBlock } from '@deepseek-ai/dsh-llm'
 import {
   appendDelegatedPolicyOverrides,
@@ -52,6 +53,10 @@ function toStopReason(reason: TurnEndReason | undefined): SubagentStopReason {
       return 'max-tokens'
     case 'aborted':
       return 'aborted'
+    // A pre-step rejection discarded the claimed prompt: the task was
+    // declined, and the caller must not read the run as done.
+    case 'blocked':
+      return 'refusal'
     case 'error':
     case 'interrupted':
     default:
@@ -207,7 +212,11 @@ function readResult(
   structured?: { captured?: { value: unknown } | undefined },
 ): SubagentResult {
   const own = child.session.events.slice(boundary)
-  const lastEnd = findLastMessageTurnEnd(own)
+  // `droppedUnrun` is deliberately unread: a one-shot prompt is claimed by its
+  // awaited first turn almost immediately, and the owner's own teardown is the
+  // `cancelled` flag below. A cancellation with no accounting turn resolves
+  // `error` through `toStopReason(undefined)`, which never overstates success.
+  const lastEnd = foldConsumedWork(own).end
   // The seam's canonical selection rule; a partial answer survives cancel and truncation.
   const output: ContentBlock[] = finalAssistantOutput(own) ?? []
   const recorded = toStopReason(lastEnd?.data.reason)
diff --git a/packages/subagent/subagent-inprocess/tests/subagent-inprocess.spec.ts b/packages/subagent/subagent-inprocess/tests/subagent-inprocess.spec.ts
index 4b9041ebf9..bc25a79c23 100644
--- a/packages/subagent/subagent-inprocess/tests/subagent-inprocess.spec.ts
+++ b/packages/subagent/subagent-inprocess/tests/subagent-inprocess.spec.ts
@@ -82,6 +82,20 @@ describe('startInProcessRun', () => {
     await run.dispose()
   })
 
+  it('reports a prompt a pre-step rejection discarded as refusal, not completion', async () => {
+    const { ctx, parent } = await setup([])
+    // A UserPromptSubmit deny or a policy plugin: the child claims its prompt,
+    // the rejection discards it, and the turn closes `blocked` with no step.
+    ctx.on('agent/pre-step', async ({ agent: subject }, next) => {
+      if (subject === parent) return next()
+      return { kind: 'reject' as const }
+    })
+
+    const run = await startInProcessRun(request(parent), {})
+    await expect(run.result).resolves.toMatchObject({ stopReason: 'refusal' })
+    await run.dispose()
+  })
+
   it('does not add a final durability checkpoint to a foreground run', async () => {
     const { ctx, parent } = await setup([textResponse('driver answer')])
     let flushes = 0
diff --git a/packages/subagent/subagent/README.i18n.yaml b/packages/subagent/subagent/README.i18n.yaml
index f3df2a0e62..c42fc3fa16 100644
--- a/packages/subagent/subagent/README.i18n.yaml
+++ b/packages/subagent/subagent/README.i18n.yaml
@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write packages/subagent/subagent/README.md
-README.md: b9f65befab71edc0685f1a6cca767c803afcb76c
-README.zh.md: f58975815742ccbe790ca1cf096449831144a2f3
+README.md: 27b01188cc1663d9cf59d3d24c91b3dbb35bea5a
+README.zh.md: e28a8794cfb227dd475ea2b4c42605a90ec7fa1d
diff --git a/packages/subagent/subagent/README.md b/packages/subagent/subagent/README.md
index b9f65befab..27b01188cc 100644
--- a/packages/subagent/subagent/README.md
+++ b/packages/subagent/subagent/README.md
@@ -76,6 +76,14 @@ The manager derives three internal residency conditions from Agent quiescence an
 
 The manager reserves the child identity, resolves the durable descriptor, calls `ctx.agents.create()` (or `ctx.agents.resume()` for cold resume) through a private activation-owner scope, installs the returned `AgentHandle` in the Activation, establishes any continuable-parent ownership, and then submits the prompt. Cold resume never dispatches through a provider because the persisted Session already holds the initial prefix and the folded descriptor is the whole reconstruction input.
 
+### Settlement delivery
+
+When a resident Activation settles, the manager tells the child's durable direct parent, in the parent's own turn stream, that the child produced everything it is going to. Delivery is unconditional for every child whose id a caller actually received: it does not consider whether the child called `report`, because the endings that most need an account — a token ceiling, a model failure, cancellation, teardown — are exactly the ones where the child never got to choose. A materialization rolled back before its first accepted message stays silent, since that caller was told the child was not established. The message carries the epoch's stop reason, its final assistant content when it produced any, and durable provenance `{ kind: 'subagent-settled', form: 'notice', senderSessionId: <child-id> }` — a different source kind from a child-authored `subagent-report`, so a transcript never credits the child with words the runtime wrote.
+
+Two ordering rules make the delivery reliable rather than lucky, and both are why this belongs to the manager instead of an external `subagent/end` listener. First, the send happens **before** the child's ownership release, while the parent still counts the child and is therefore structurally unable to be judged settled. Second, a parent that is itself a resident Activation receives the message through the same waking-admission accounting as a report, so the window between the synchronous send and the microtask that admits it is not mistaken for quiescence — `Agent.status` folds context maintenance into `idle`, and a waking send behind maintenance only arms a deferred wake. Without either rule the parent can be disposed with the notice still in an inbox that `cancel()` clears, which loses it silently.
+
+An idle parent receives the notice as one ordinary later turn. A busy parent is steered into its nearest step boundary instead, so several children settling together cost one step rather than one turn each; steering rather than injecting also means a driver that retires between the status read and the send still claims the message. A parent whose own lineage is already draining receives the notice by injection, with no wake at all: `Agent.followup()` on a quiescent parent starts a turn and `cancel()` does not arm against a later one, so waking during teardown would spend a model request on an Agent its host is about to dispose — once per tree layer, since each layer's notice then wakes the layer above it. The injected message reaches a parent that is still reading its inbox, and the log records the account either way, but it does not outlive that parent's own disposal: `AgentHandle.dispose()` is a `keepInbox: false` cancel, which durably cancels an unclaimed notice. A resumed parent therefore has no pending notice to read: `list_agents` tells it which children exist and whether each is live or stored, while the outcome itself stays in the child's own Session, which a `send_message` reaches by resuming that child. A parent that has left the registry is not an error: the notice is dropped and the child's own Session remains the durable record. Delivery never blocks or fails teardown — a rejected send is logged, because retaining a child to retry a notice would pin its whole ancestry in `waiting` forever.
+
 A continuation-managed parent Activation records each child Session id in an `ownedChildren` set before the child can run and disposes only after every owned child Activation completes `AgentHandle` disposal (child-first). Teardown propagates Agent cancellation top-down before awaiting slow descendants, while handle release remains child-first. Top-level and other non-continuation Agents have no Activation and stay outside this waiting graph. Final settlement awaits a best-effort `ctx.sessions.flush(child.session)` before handle disposal. A listener rejection is logged without failing the Activation because listener participation does not identify a persistence backend; the persisted state may therefore be missing or stale on resume.
 
 ## Lifecycle events
@@ -94,17 +102,31 @@ When `ctx.sessionProjections` is available, the service registers two projection
 
 ## Collection model
 
-The model-facing tool collects synchronously by default: it awaits the child result and disposes the run before returning. One-shot background delegation registers a plain Task in the tool, whose generic status, collection, and cancellation tools own later interaction, and persists its model-supplied `description` as the optional display label. Continuable background delegation calls `ctx.subagents.startContinuable()` and returns only the durable child id; the child owns its own turns from inbox acceptance, so there is no Task and no result promise — a caller sends later work with the `send_message` follow-up tool, and `interrupt()` stops only the current turn without disposing the child, and the durable child Session remains the source of the child's detailed output. The continuation manager exists only while `ctx.agents` is available, and session persistence is resolved per continuation operation. Independently, `listChildren()` enumerates the live-preferred merge of the live session store and optional session persistence — live-only when persistence is absent, since a cold child cannot be resumed then either — and serves each child's durable mode/label from the registered `subagent` projection unit: the registry's watermark snapshot for a live child; for a cold one, a durable projection-cache row when it serves an own-suffix identity — its `seq` gate proves the value postdates the fork seed, where a child's own descriptor is immutable once appended — else one bounded-concurrency persistence inspection folded through the registry, whose result must still name the enumerated lifecycle (a re-published id degrades to a `corrupt` diagnostic). A throwing cache read renders no verdict — the cache is derived data — and silently falls through to that authoritative re-fold. The projection fold is the single classification authority; listing parses no descriptor itself. A served identity produces a child row; a settled candidate whose fold served no identity is a `corrupt` diagnostic, a failed inspection is a transient `unavailable` retried on the next listing, and a running candidate without an identity yet is omitted (the creation window before its descriptor is appended). It never consults the continuation manager, Agent registrations, Activations, or providers. Each child row derives its read-time `hasChildren` hint from merged headers carrying durable `origin: 'subagent'`; it does not read descendant event logs, and the descriptor-backed child catalog remains authoritative when expanded. Service consumers such as a UI can retain both modes and choose a fallback for an unlabeled one-shot child; the model-facing `list_agents` tool projects only `continuable` entries and refines status through the live Agent registry (`running`/`idle`/`complete`) and walks `listDescendants()` for its `descendants` scope. The listing forwards the caller's signal to every persistence read, checks cancellation around each of those awaits, and reports every observed abort as `SubagentError` code `CANCELLED`; an unmounted projection registry fails loud with `SUBAGENT_CONTROL_PROJECTIONS_UNAVAILABLE`, and a missing session store with `SUBAGENT_CONTROL_SESSION_STORE_UNAVAILABLE`. See the [background subagent tasks Agent Note](../../../.agents/notes/implemented/feature/2026-07-08-background-subagent-tasks.md), the [continuable background subagents Agent Note](../../../.agents/notes/implemented/feature/2026-07-21-continuable-background-subagents.md), the [durable catalog Agent Note](../../../.agents/notes/implemented/feature/2026-07-22-durable-subagent-catalog-and-list-agents.md), the [merged-service Agent Note](../../../.agents/notes/implemented/simplification/2026-07-26-merge-subagent-control-service.md), the [capability-seam Agent Note](../../../.agents/notes/implemented/feature/2026-06-21-subagent-capability-seam.md), and `src/types.ts` for the complete contracts.
+The model-facing tool collects synchronously by default: it awaits the child result and disposes the run before returning. One-shot background delegation registers a plain Task in the tool, whose generic status, collection, and cancellation tools own later interaction, and persists its model-supplied `description` as the optional display label. Continuable background delegation calls `ctx.subagents.startContinuable()` and returns only the durable child id; the child owns its own turns from inbox acceptance, so there is no Task and no result promise — a caller sends later work with the `send_message` follow-up tool, and `interrupt()` stops only the current turn without disposing the child, and the durable child Session remains the source of the child's detailed output. The continuation manager exists only while `ctx.agents` is available, and session persistence is resolved per continuation operation. Independently, `listChildren()` enumerates the live-preferred merge of the live session store and optional session persistence — live-only when persistence is absent, since a cold child cannot be resumed then either — and serves each child's durable mode/label from the registered `subagent` projection unit: the registry's watermark snapshot for a live child; for a cold one, a durable projection-cache row when it serves an own-suffix identity — its `seq` gate proves the value postdates the fork seed, where a child's own descriptor is immutable once appended — else one bounded-concurrency persistence inspection folded through the registry, whose result must still name the enumerated lifecycle (a re-published id degrades to a `corrupt` diagnostic). A throwing cache read renders no verdict — the cache is derived data — and silently falls through to that authoritative re-fold. The projection fold is the single classification authority; listing parses no descriptor itself. A served identity produces a child row; a settled candidate whose fold served no identity is a `corrupt` diagnostic, a failed inspection is a transient `unavailable` retried on the next listing, and a running candidate without an identity yet is omitted (the creation window before its descriptor is appended). It never consults the continuation manager, Agent registrations, Activations, or providers. Each child row derives its read-time `hasChildren` hint from merged headers carrying durable `origin: 'subagent'`; it does not read descendant event logs, and the descriptor-backed child catalog remains authoritative when expanded. Service consumers such as a UI can retain both modes and choose a fallback for an unlabeled one-shot child; the model-facing `list_agents` tool projects only `continuable` entries and refines status through the live Agent registry and maps storage-only to its resumable-not-terminal `ready` (`running`/`idle`/`ready`) and walks `listDescendants()` for its `descendants` scope. The listing forwards the caller's signal to every persistence read, checks cancellation around each of those awaits, and reports every observed abort as `SubagentError` code `CANCELLED`; an unmounted projection registry fails loud with `SUBAGENT_CONTROL_PROJECTIONS_UNAVAILABLE`, and a missing session store with `SUBAGENT_CONTROL_SESSION_STORE_UNAVAILABLE`. See the [background subagent tasks Agent Note](../../../.agents/notes/implemented/feature/2026-07-08-background-subagent-tasks.md), the [continuable background subagents Agent Note](../../../.agents/notes/implemented/feature/2026-07-21-continuable-background-subagents.md), the [durable catalog Agent Note](../../../.agents/notes/implemented/feature/2026-07-22-durable-subagent-catalog-and-list-agents.md), the [merged-service Agent Note](../../../.agents/notes/implemented/simplification/2026-07-26-merge-subagent-control-service.md), the [capability-seam Agent Note](../../../.agents/notes/implemented/feature/2026-06-21-subagent-capability-seam.md), and `src/types.ts` for the complete contracts.
 
 Continuable Activations await a best-effort final session flush without treating listener participation as durability confirmation. One-shot runs retain best-effort session checkpointing, so a completed one-shot child is discoverable after disposal only when its session actually reached persistence; the service does not invent a catalog entry from Task history when that checkpoint is absent.
 
 ## Model Experience
 
+### Settlement notice
+
+#### What the model sees
+
+One user-role parent message opening with the outcome — `Background subagent <child-id> finished and will do no further work unless you send it more.`, or the matching line for a child that was stopped, ran out of room, declined, or failed — followed by `Its closing message:` and the child's final assistant content, or `It left no closing message.` when it produced none. This is the service's only direct parent-side contribution; delegation schemas, parent continuation and discovery, and the child-scoped `report` belong to `dsh-tool-subagent`, `dsh-tool-subagent-control`, and `dsh-tool-subagent-report`.
+
+#### Token effect
+
+One notice per settled Activation in the parent's request, sized by the child's final message. A child that both reports and settles costs the parent both.
+
+#### KV Cache effect
+
+Append-only in the parent: the notice follows its reusable request prefix. Reaching an idle parent starts one independent model request; reaching a busy one does not.
+
 ### Child delegation-scope statement
 
 #### What the model sees
 
-Every in-process child's runtime-context snapshot carries the `subagent:delegation` statement below, after the sandbox-policy and approval-policy sentences; parent-side rendering stays with `dsh-tool-subagent` (delegation schemas), `dsh-tool-subagent-control` (continuation and discovery), and `dsh-tool-subagent-report` (the child-scoped `report`).
+Every in-process child's runtime-context snapshot carries the `subagent:delegation` statement below, after the sandbox-policy and approval-policy sentences.
 
 ##### The delegation-scope statement
 
diff --git a/packages/subagent/subagent/README.zh.md b/packages/subagent/subagent/README.zh.md
index f589758157..e28a8794cf 100644
--- a/packages/subagent/subagent/README.zh.md
+++ b/packages/subagent/subagent/README.zh.md
@@ -76,6 +76,14 @@ subagent seam 允许一个 agent（智能体）通过具名提供方把工作委
 
 管理器预留子 agent 身份、解析持久化描述符，通过私有的 activation-owner 作用域调用 `ctx.agents.create()`（冷恢复时为 `ctx.agents.resume()`），把返回的 `AgentHandle` 安装到 Activation 中，建立任何可继续父级所有权，然后提交提示词。冷恢复绝不通过提供方分发，因为持久化 Session 已持有初始前缀，折叠后的描述符即是全部重建输入。
 
+### 结算投递
+
+当一个驻留 Activation 结算时，管理器会在父级自身的轮次流中告知该子级持久化的直接父级：这个子级已经产出它将产出的全部内容。对每个调用方真正拿到过 id 的子级，这条投递都是无条件的：它不考虑该子级是否调用过 `report`，因为最需要这条投递的结束方式——token 上限、模型失败、取消、拆卸——恰恰是子级根本没有机会选择的那些。在第一条消息被接受之前就回滚的物化保持静默，因为那位调用方已被告知该子级未建立。消息会携带该 epoch 的终止原因、它产出过的最终 assistant 内容，以及持久化来源 `{ kind: 'subagent-settled', form: 'notice', senderSessionId: <child-id> }`——与子级自撰的 `subagent-report` 是不同的来源 kind，因此 transcript（文本记录）绝不会把运行时写下的话算到子级头上。
+
+有两条顺序规则让这条投递可靠而非侥幸，它们也正是这件事属于管理器而非外部 `subagent/end` listener 的原因。第一，发送发生在子级所有权释放**之前**，此时父级仍然计入该子级，因此在结构上不可能被判定为已结算。第二，本身就是驻留 Activation 的父级会通过与 report 相同的唤醒准入记账接收该消息，因此同步发送与承认它的那个 microtask 之间的窗口不会被误判为静止——`Agent.status` 会把上下文维护折叠成 `idle`，而维护期间的唤醒发送只会预置一次延后唤醒。缺少其中任一条规则，父级都可能在通知仍留在 inbox 时被 dispose，而 `cancel()` 会清空该 inbox，于是通知被静默丢失。
+
+空闲父级会以一个普通的后续轮次收到该通知。繁忙父级则被 steer 到其最近的 step 边界，因此同时结算的多个子级只消耗一个 step，而不是各自一个轮次；采用 steer 而非 inject 还意味着：即便驱动在状态读取与发送之间退出，该消息仍会被认领。若父级自身所在的谱系已在 draining，则该通知改为 inject 送达，完全不唤醒：对静息父级调用 `Agent.followup()` 会开启一个轮次，而 `cancel()` 不会对之后的轮次设防，因此在拆卸期间唤醒，会在宿主即将 dispose 的 Agent 上白花一次模型请求——而且每层树各一次，因为每层自己的通知又会唤醒它上面那层。被 inject 的消息会送达仍在读取自身 inbox 的父级，而无论如何日志都会记录这份记账；但它不会比该父级自身的 dispose 活得更久：`AgentHandle.dispose()` 是一次 `keepInbox: false` 的 cancel，会持久地取消尚未被认领的通知。因此 resume 后的父级没有待处理通知可读：`list_agents` 只告诉它有哪些子级、各自是在线还是仅存于存储；结局本身留在子级自己的 Session 里，一次 `send_message` 会通过 resume 该子级把它取回。已离开注册表的父级不算错误：通知被丢弃，子级自身的 Session 仍是持久记录。投递绝不会阻塞或使拆卸失败——发送被拒只会记录日志，因为为了重试一条通知而保留子级，会把它的整条祖先链永久钉在 `waiting` 上。
+
 受继续执行管理的父级 Activation 会在子 agent 能够运行之前，把每个子 agent 的 Session id 记录到 `ownedChildren` 集合中，并且只有在每个所拥有的子 agent Activation 完成 `AgentHandle` dispose 之后才会 dispose（子先于父）。拆卸会先自顶向下传播 Agent 取消，再等待缓慢的后代，而 handle 释放仍保持 child-first。顶层及其他非继续执行的 Agent 没有 Activation，处于该等待图之外。最终结算会在 dispose handle 前等待 best-effort 的 `ctx.sessions.flush(child.session)`。listener rejection 会被记录，但不会使 Activation 失败，因为 listener 是否参与无法标识持久化后端；因此，恢复时持久化状态可能缺失或陈旧。
 
 ## 生命周期事件
@@ -94,17 +102,31 @@ subagent seam 允许一个 agent（智能体）通过具名提供方把工作委
 
 ## 收集模型
 
-面向模型的工具默认同步收集：先等待子 agent 结果，再 dispose 运行，然后才返回。一次性后台委派会在工具中注册普通 Task，其通用状态、收集和取消工具负责后续交互，并将模型提供的 `description` 持久化为可选显示标签。可继续后台委派会调用 `ctx.subagents.startContinuable()`，只返回持久化子 agent id；子 agent 自 inbox 接受起就拥有自己的轮次，因此没有 Task、也没有结果 promise——调用方通过 `send_message` 后续操作工具发送后续工作，`interrupt()` 只停止当前轮次而不 dispose 子 agent，而持久化子 agent Session 仍是子 agent 详细输出的来源。只有 `ctx.agents` 可用时，继续执行管理器才会存在，而会话持久化按每项继续执行操作解析。与此独立，`listChildren()` 枚举在线会话存储与可选会话持久化的在线优先合并——持久化缺席时仅枚举在线 child，因为那时冷 child 本就无法恢复——并由已注册的 `subagent` 投影单元供给每个 child 的持久化模式与标签：在线 child 取注册表的水位快照；冷 child 先取可选投影缓存的持久化行，且仅当其 `seq` 门证明该值折叠自 child 自身后缀（fork 种子之后——自有描述符一经追加即不可变）才直接采用，否则经一次有界并发的持久化 inspect 再经注册表折叠，且 inspect 结果必须仍指向枚举时的生命周期（同 id 被重新发布的会话降级为 `corrupt` diagnostic）。缓存读取抛错不产生判决——缓存是派生数据——静默落到该权威重折。投影折叠是唯一的分类权威；列表自身不解析任何描述符。取得身份值即产出 child 行；已定局而折叠未产出身份的候选是 `corrupt` diagnostic，inspect 失败是瞬时的 `unavailable`（下次列表重试），运行中而暂无身份值的候选整行省略（描述符尚未追加的创建窗口）。它不查询继续执行管理器、Agent 注册信息、Activation 或提供方。每个 child 行都会根据合并结果中携带持久化 `origin: 'subagent'` 的 header 派生读取时的 `hasChildren` 提示；它不会读取后代事件日志，展开后仍以描述符支撑的 child 目录为权威依据。UI 等服务消费方可以保留两种模式，并为无标签的一次性 child 选择回退展示；面向模型的 `list_agents` 工具只投影 `continuable` 条目，通过在线 Agent 注册表细化状态（`running`／`idle`／`complete`），并在 `descendants` scope 下遍历 `listDescendants()`。列表操作会把调用方的取消信号转发到每次持久化读取，在这些 await 前后检查取消，并将每次检测到的中止报告为 `SubagentError` 错误码 `CANCELLED`；投影注册表未挂载则以 `SUBAGENT_CONTROL_PROJECTIONS_UNAVAILABLE` 响亮失败，会话存储缺失则以 `SUBAGENT_CONTROL_SESSION_STORE_UNAVAILABLE` 响亮失败。完整约定见[后台 subagent 任务 Agent Note](../../../.agents/notes/implemented/feature/2026-07-08-background-subagent-tasks.md)、[可继续后台 subagent Agent Note](../../../.agents/notes/implemented/feature/2026-07-21-continuable-background-subagents.md)、[持久化目录 Agent Note](../../../.agents/notes/implemented/feature/2026-07-22-durable-subagent-catalog-and-list-agents.md)、[服务合并 Agent Note](../../../.agents/notes/implemented/simplification/2026-07-26-merge-subagent-control-service.md)、[能力 seam Agent Note](../../../.agents/notes/implemented/feature/2026-06-21-subagent-capability-seam.md)和 `src/types.ts`。
+面向模型的工具默认同步收集：先等待子 agent 结果，再 dispose 运行，然后才返回。一次性后台委派会在工具中注册普通 Task，其通用状态、收集和取消工具负责后续交互，并将模型提供的 `description` 持久化为可选显示标签。可继续后台委派会调用 `ctx.subagents.startContinuable()`，只返回持久化子 agent id；子 agent 自 inbox 接受起就拥有自己的轮次，因此没有 Task、也没有结果 promise——调用方通过 `send_message` 后续操作工具发送后续工作，`interrupt()` 只停止当前轮次而不 dispose 子 agent，而持久化子 agent Session 仍是子 agent 详细输出的来源。只有 `ctx.agents` 可用时，继续执行管理器才会存在，而会话持久化按每项继续执行操作解析。与此独立，`listChildren()` 枚举在线会话存储与可选会话持久化的在线优先合并——持久化缺席时仅枚举在线 child，因为那时冷 child 本就无法恢复——并由已注册的 `subagent` 投影单元供给每个 child 的持久化模式与标签：在线 child 取注册表的水位快照；冷 child 先取可选投影缓存的持久化行，且仅当其 `seq` 门证明该值折叠自 child 自身后缀（fork 种子之后——自有描述符一经追加即不可变）才直接采用，否则经一次有界并发的持久化 inspect 再经注册表折叠，且 inspect 结果必须仍指向枚举时的生命周期（同 id 被重新发布的会话降级为 `corrupt` diagnostic）。缓存读取抛错不产生判决——缓存是派生数据——静默落到该权威重折。投影折叠是唯一的分类权威；列表自身不解析任何描述符。取得身份值即产出 child 行；已定局而折叠未产出身份的候选是 `corrupt` diagnostic，inspect 失败是瞬时的 `unavailable`（下次列表重试），运行中而暂无身份值的候选整行省略（描述符尚未追加的创建窗口）。它不查询继续执行管理器、Agent 注册信息、Activation 或提供方。每个 child 行都会根据合并结果中携带持久化 `origin: 'subagent'` 的 header 派生读取时的 `hasChildren` 提示；它不会读取后代事件日志，展开后仍以描述符支撑的 child 目录为权威依据。UI 等服务消费方可以保留两种模式，并为无标签的一次性 child 选择回退展示；面向模型的 `list_agents` 工具只投影 `continuable` 条目，通过在线 Agent 注册表细化状态，并把仅存于存储的状态映射为可恢复而非终态的 `ready`（`running`／`idle`／`ready`），并在 `descendants` scope 下遍历 `listDescendants()`。列表操作会把调用方的取消信号转发到每次持久化读取，在这些 await 前后检查取消，并将每次检测到的中止报告为 `SubagentError` 错误码 `CANCELLED`；投影注册表未挂载则以 `SUBAGENT_CONTROL_PROJECTIONS_UNAVAILABLE` 响亮失败，会话存储缺失则以 `SUBAGENT_CONTROL_SESSION_STORE_UNAVAILABLE` 响亮失败。完整约定见[后台 subagent 任务 Agent Note](../../../.agents/notes/implemented/feature/2026-07-08-background-subagent-tasks.md)、[可继续后台 subagent Agent Note](../../../.agents/notes/implemented/feature/2026-07-21-continuable-background-subagents.md)、[持久化目录 Agent Note](../../../.agents/notes/implemented/feature/2026-07-22-durable-subagent-catalog-and-list-agents.md)、[服务合并 Agent Note](../../../.agents/notes/implemented/simplification/2026-07-26-merge-subagent-control-service.md)、[能力 seam Agent Note](../../../.agents/notes/implemented/feature/2026-06-21-subagent-capability-seam.md)和 `src/types.ts`。
 
 可继续 Activation 会等待 best-effort 的最终会话 flush，但不会把 listener 参与视为持久性确认。一次性运行保留尽力执行的会话检查点，因此已完成的一次性 child 只有在其会话确实进入持久化存储时，才可在 dispose 后继续被发现；如果该检查点缺失，服务不会根据 Task 历史虚构目录条目。
 
 ## 模型体验
 
+### 结算通知
+
+#### 模型看到的内容
+
+一条用户角色的父级消息，开头是结果本身——`Background subagent <child-id> finished and will do no further work unless you send it more.`，或子级被停止、耗尽额度、拒绝任务或失败时的对应句子——随后是 `Its closing message:` 与子级的最终 assistant 内容；若子级没有产出内容，则是 `It left no closing message.`。这是本服务面向父级的唯一直接贡献；委派 schema、父级延续与发现以及子级作用域的 `report` 分别归 `dsh-tool-subagent`、`dsh-tool-subagent-control` 和 `dsh-tool-subagent-report` 所有。
+
+#### Token 影响
+
+父级请求中，每个已结算的 Activation 一条通知，长度取决于子级的最终消息。既上报又结算的子级会让父级同时支付两份。
+
+#### KV Cache 影响
+
+在父级中仅追加：通知位于其可复用请求前缀之后。到达空闲父级会启动一次独立的模型请求，到达繁忙父级则不会。
+
 ### 子级委派范围声明
 
 #### 模型看到的内容
 
-每个进程内子 agent 的运行时上下文快照都携带下方的 `subagent:delegation` 声明，位于沙箱策略与审批策略语句之后；父级侧的渲染仍归 `dsh-tool-subagent`（委派 schema）、`dsh-tool-subagent-control`（延续与发现）和 `dsh-tool-subagent-report`（子级作用域的 `report`）所有。
+每个进程内子 agent 的运行时上下文快照都携带下方的 `subagent:delegation` 声明，位于沙箱策略与审批策略语句之后。
 
 ##### 委派范围声明
 
diff --git a/packages/subagent/subagent/src/continuation.ts b/packages/subagent/subagent/src/continuation.ts
index 871ec6efdd..073708f4bc 100644
--- a/packages/subagent/subagent/src/continuation.ts
+++ b/packages/subagent/subagent/src/continuation.ts
@@ -1,7 +1,8 @@
 /**
  * Internal continuable-subagent manager: stable child ids, descriptor
  * persistence, activation admission, the live ownership graph, cold resume,
- * and child-first disposal behind `ctx.subagents`.
+ * child-first disposal, and settlement delivery to the parent, behind
+ * `ctx.subagents`.
  *
  * A continuable child has one durable Session and at most one process-local
  * {@link Activation} — one residency epoch for a reconstructed child Agent. An
@@ -11,6 +12,12 @@
  * residency while the Agent loop owns all turn ordering and execution. No
  * continuable path creates a Task or an intermediate result-bearing wrapper.
  *
+ * Because residency is this manager's alone to end, telling the parent that a
+ * child settled is its job too. An external `subagent/end` listener cannot do
+ * it correctly: that payload names no parent, the child handle is already
+ * disposed by then, and the release that wakes the parent's own settlement
+ * watcher has already run. See {@link SubagentContinuationManager.notifySettlement}.
+ *
  * @module @deepseek-ai/dsh-subagent
  */
 
@@ -23,7 +30,7 @@ import type {
   AgentSetupCommit,
   CreateAgentOptions,
 } from '@deepseek-ai/dsh-agent'
-import { createUserMessage, errorChain } from '@deepseek-ai/dsh-llm'
+import { boundContextSummary, createUserMessage, errorChain } from '@deepseek-ai/dsh-llm'
 import type { ContentBlock, MessageId, MessageSource } from '@deepseek-ai/dsh-llm'
 import { SessionId } from '@deepseek-ai/dsh-session'
 import type { SessionEvent } from '@deepseek-ai/dsh-session'
@@ -42,8 +49,8 @@ import {
 import type { DelegatedPolicyOverrides } from './child-agent.ts'
 import { assertSubagentMaxDepth } from './depth.ts'
 import { seedDescriptorTurn } from './descriptor-seed.ts'
-import type { ContinuableCreateRequest, ContinuableCreateSpec, SubagentStartRequest } from './types.ts'
-import type { ActivationObserver } from './lifecycle.ts'
+import type { ContinuableCreateRequest, ContinuableCreateSpec, SubagentResult, SubagentStartRequest } from './types.ts'
+import type { ActivationObserver, ActivationTerminal } from './lifecycle.ts'
 import { SubagentError } from './error.ts'
 import type SubagentActivationSetupRegistry from './activation-setup-registry.ts'
 
@@ -65,10 +72,28 @@ export interface SubagentReportMessageSource {
   readonly senderSessionId: SessionId
 }
 
+/**
+ * Durable attribution for the runtime's own account of a continuable child
+ * settling. Deliberately a different kind from
+ * {@link SubagentReportMessageSource}: a report is content the child chose,
+ * while this message is the manager stating what became of the child, and a
+ * transcript that merged them would credit the child with words it never wrote.
+ */
+export interface SubagentSettledMessageSource {
+  readonly kind: 'subagent-settled'
+  /** A runtime account shown without expanding the row (`notice` context form). */
+  readonly form: 'notice'
+  /** One-line account of how the child ended. */
+  readonly summary: string
+  /** Session id of the child that settled. */
+  readonly senderSessionId: SessionId
+}
+
 declare module '@deepseek-ai/dsh-llm' {
   interface MessageSourceMap {
     coordinator: CoordinatorMessageSource
     'subagent-report': SubagentReportMessageSource
+    'subagent-settled': SubagentSettledMessageSource
   }
 }
 
@@ -166,6 +191,13 @@ interface ContinuationHost {
 interface Activation {
   /** The durable child this Activation is an epoch of. */
   readonly childId: SessionId
+  /**
+   * The durable direct parent, stored because settlement delivery must resolve
+   * that parent after the child handle is gone. {@link ancestry} cannot answer
+   * it: a `WeakSet` is not enumerable, and the child's own header is only
+   * reachable through a handle disposal has already released.
+   */
+  readonly parentSession: SessionId
   /** The provider name recorded in the durable descriptor. */
   readonly provider: string
   /** The retained live Agent handle, disposed exactly once at settlement. */
@@ -197,6 +229,12 @@ interface Activation {
    * microtask that admits it, so settlement must not treat that gap as quiet.
    */
   readonly accepted: Set<MessageId>
+  /**
+   * Whether any delivery to this child was ever accepted. A materialization
+   * rolled back before its first acceptance is a child the caller was told does
+   * not exist, so its teardown owes the parent no settlement account.
+   */
+  announced: boolean
   /** Renewed whenever a settlement watcher must re-observe quiescence. */
   poke: PromiseWithResolvers<void>
 }
@@ -243,6 +281,36 @@ function disposalOf(activation: Activation): Promise<void> | undefined {
   return activation.disposal
 }
 
+/**
+ * One line telling a parent that a background child is finished and why, in
+ * the parent's own task vocabulary.
+ * @param childId - the durable child the parent knows by id.
+ * @param stopReason - how the child's last ordinary turn ended.
+ * @returns the model-facing opening line of the settlement notice.
+ */
+function settlementSummary(childId: SessionId, stopReason: SubagentResult['stopReason']): string {
+  const subject = `Background subagent ${childId}`
+  switch (stopReason) {
+    case 'completed':
+      return `${subject} finished and will do no further work unless you send it more.`
+    case 'aborted':
+      return `${subject} was stopped before it finished.`
+    case 'max-tokens':
+      return `${subject} ran out of room before it finished.`
+    // A pre-step rejection — a hook deny, a policy plugin — discarded input
+    // the child had claimed, so the parent must not treat the task as done.
+    case 'refusal':
+      return `${subject} declined the task.`
+    case 'error':
+      return `${subject} failed before it finished.`
+    /* v8 ignore next 4 -- `SubagentResult['stopReason']` is merge-extensible, so this arm
+     * needs a backend that adds a variant; an unnameable ending is reported as unfinished
+     * rather than silently as success. */
+    default:
+      return `${subject} ended abnormally (${String(stopReason)}) before it finished.`
+  }
+}
+
 /** Whether one settlement attempt opened the disposal transaction. */
 type SettlementAttempt =
   | { readonly settling: false }
@@ -576,19 +644,36 @@ export class SubagentContinuationManager {
         senderSessionId: activation.childId,
       },
     })
-    const parentActivation = this.activations.get(parent.id)
-    if (delivery === 'wakeup'
-      && parentActivation !== undefined
-      && parentActivation.handle.agent === parent) {
-      this.admitWaking(parentActivation, message.id, () => {
-        this.sendReport(parent, message, delivery)
-      })
+    if (delivery === 'wakeup') {
+      this.sendWaking(parent, message, () => { this.sendReport(parent, message, delivery) })
     } else {
       this.sendReport(parent, message, delivery)
     }
     return message.id
   }
 
+  /**
+   * Perform one waking send to a parent, accounted against that parent's own
+   * Activation when it has one. Registering the id before the send is what
+   * keeps a continuation-managed parent from being judged quiescent in the
+   * window between `followup()` and the microtask that admits it.
+   * @param parent - the exact live parent receiving the waking message.
+   * @param message - the message whose id is accounted.
+   * @param send - the synchronous waking send to perform.
+   */
+  private sendWaking(
+    parent: Agent,
+    message: ReturnType<typeof createUserMessage>,
+    send: () => void,
+  ): void {
+    const parentActivation = this.activations.get(parent.id)
+    if (parentActivation !== undefined && parentActivation.handle.agent === parent) {
+      this.admitWaking(parentActivation, message.id, send)
+    } else {
+      send()
+    }
+  }
+
   /** Send one report while translating only the parent's own rejection. */
   private sendReport(
     parent: Agent,
@@ -745,23 +830,32 @@ export class SubagentContinuationManager {
     return lineage
   }
 
-  /** Reject new admission once the manager or this exact parent tree began draining. */
-  private assertAdmitting(agent: Agent): void {
-    if (this.draining) {
-      throw new SubagentError(
-        'continuable subagents are draining; the operation was not admitted',
-        'DRAINING',
-      )
-    }
+  /**
+   * The teardown that closed continuable admission for this agent's lineage.
+   * `'manager'` is the whole manager draining; an Agent is the exact scoped root
+   * whose forest is closing.
+   * @param agent - the agent whose lineage is tested.
+   * @returns the closing teardown, or `undefined` while admission is open.
+   */
+  private closingTeardownFor(agent: Agent): Agent | 'manager' | undefined {
+    if (this.draining) return 'manager'
     const lineage = this.liveLineage(agent)
     for (const [root, members] of this.closingScopes) {
-      if (members.has(agent) || lineage.includes(root)) {
-        throw new SubagentError(
-          `continuable subagents below parent "${root.id}" are draining; the operation was not admitted`,
-          'DRAINING',
-        )
-      }
+      if (members.has(agent) || lineage.includes(root)) return root
     }
+    return undefined
+  }
+
+  /** Reject new admission once the manager or this exact parent tree began draining. */
+  private assertAdmitting(agent: Agent): void {
+    const closing = this.closingTeardownFor(agent)
+    if (closing === undefined) return
+    throw new SubagentError(
+      closing === 'manager'
+        ? 'continuable subagents are draining; the operation was not admitted'
+        : `continuable subagents below parent "${closing.id}" are draining; the operation was not admitted`,
+      'DRAINING',
+    )
   }
 
   /**
@@ -930,6 +1024,10 @@ export class SubagentContinuationManager {
 
     const activation: Activation = {
       childId,
+      // The durable lineage, not merely the caller: creation stamps this same
+      // agent into the child's header, and cold resume authorized it against
+      // the persisted header before materializing.
+      parentSession: parent.id,
       provider,
       handle,
       ancestry: new WeakSet([handle.agent, ...parentLineage]),
@@ -937,6 +1035,7 @@ export class SubagentContinuationManager {
       observer,
       disposal: undefined,
       accepted: new Set(),
+      announced: false,
       poke: Promise.withResolvers<void>(),
     }
     // After transfer, any failure must dispose the created handle, remove the
@@ -1038,9 +1137,13 @@ export class SubagentContinuationManager {
     // establish it before the message can enter the child's inbox.
     this.acquireOwnership(parent, activation.childId)
     const message = createUserMessage({ content, source })
-    return this.admitWaking(activation, message.id, () => {
+    const accepted = this.admitWaking(activation, message.id, () => {
       activation.handle.agent.followup(message)
     })
+    // Past this point the caller has an id for this child, so its eventual
+    // settlement is something the parent is owed an account of.
+    activation.announced = true
+    return accepted
   }
 
   /**
@@ -1261,6 +1364,12 @@ export class SubagentContinuationManager {
     // makes a racing delivery wait for release rather than cold-resume into the
     // still-registered agent.
     this.activations.delete(childId)
+    // BEFORE releasing ownership, while the parent still counts this child and
+    // therefore cannot be judged settled. Delivering after the release would
+    // race a parent watcher that resumes one microtask later, finds itself
+    // childless and quiet, and disposes an Agent whose `cancel()` clears the
+    // inbox this notice is sitting in.
+    this.notifySettlement(activation, activation.observer.terminal(failure))
     // Release ownership even on failure: a retained failed child would pin its
     // ancestors in `waiting` forever.
     this.releaseOwnership(childId)
@@ -1270,6 +1379,75 @@ export class SubagentContinuationManager {
     if (failure !== undefined) throw failure
   }
 
+  /**
+   * Tell the durable direct parent that this child produced everything it is
+   * going to. Unconditional for every child the caller received an id for: it
+   * does not consider whether the child reported, because the cases that most
+   * need it — a token ceiling, a model failure, cancellation, teardown — are
+   * exactly the ones where the child never got to choose. A materialization
+   * rolled back before its first acceptance stays silent, since the caller was
+   * told that child was not established. A parent that is no longer live is not
+   * an error; the child's own Session remains the durable record either way.
+   * A parent whose own lineage is already closing receives the notice without a
+   * wake, because teardown is not a reason to start a turn.
+   *
+   * Never blocks disposal. A delivery failure is logged and dropped, because
+   * retaining a child to retry a notice would pin its whole ancestry in
+   * `waiting` forever.
+   * @param activation - the settling Activation, still owned by its parent.
+   * @param terminal - how this epoch ended, as the terminal edge will report it.
+   */
+  private notifySettlement(activation: Activation, terminal: ActivationTerminal): void {
+    if (!activation.announced) return
+    try {
+      const parent = this.ctx.agents.get(activation.parentSession)
+      if (parent === undefined) return
+      const summary = settlementSummary(activation.childId, terminal.stopReason)
+      const message = createUserMessage({
+        content: [
+          { type: 'text' as const, text: summary },
+          ...terminal.output === undefined
+            ? [{ type: 'text' as const, text: 'It left no closing message.' }]
+            : [{ type: 'text' as const, text: 'Its closing message:' }, ...terminal.output],
+        ],
+        source: {
+          kind: 'subagent-settled' as const,
+          form: 'notice' as const,
+          summary: boundContextSummary(summary),
+          senderSessionId: activation.childId,
+        },
+      })
+      // A parent whose own teardown already began must not be woken. Waking is
+      // not a queue operation: `followup()` on a quiescent Agent starts a turn,
+      // and `cancel()` does not arm against a later one, so a notice arriving
+      // during teardown would spend a model request on an Agent its host is
+      // about to dispose — once per tree layer, since each layer's own notice
+      // then wakes the layer above it. Injecting delivers to a parent still
+      // reading its inbox and records the account in the log either way; it
+      // does NOT survive that parent's own disposal, whose `keepInbox: false`
+      // cancel durably clears whatever it never claimed.
+      if (this.closingTeardownFor(parent) !== undefined) {
+        parent.inject(message)
+        return
+      }
+      // An idle parent has nothing else to look at, so it gets one ordinary
+      // turn. A busy parent is steered instead of woken: `Inbox.claim()` takes
+      // the whole next-step batch at one boundary, so several children settling
+      // together cost one step rather than one turn each. Steering rather than
+      // injecting closes the window where a driver retires between this status
+      // read and the send, which would strand the notice unclaimed.
+      this.sendWaking(parent, message, () => {
+        if (parent.status === 'idle') parent.followup(message)
+        else parent.steer(message)
+      })
+    } catch (error: unknown) {
+      this.ctx.logger.warn(
+        `subagent "${activation.childId}" settlement notice was not delivered to its parent: `
+        + errorChain(error),
+      )
+    }
+  }
+
   /**
    * Request a best-effort final session flush after the child is quiescent.
    * Listener failure is logged because flush participation cannot identify a
diff --git a/packages/subagent/subagent/src/index.ts b/packages/subagent/subagent/src/index.ts
index 36acb8de28..c31d197d25 100644
--- a/packages/subagent/subagent/src/index.ts
+++ b/packages/subagent/subagent/src/index.ts
@@ -119,6 +119,7 @@ export type {
   SubagentReportDelivery,
   SubagentReportMessageSource,
   SubagentReportOptions,
+  SubagentSettledMessageSource,
 } from './continuation.ts'
 export type { ContinuableSetupContribution } from './activation-setup-registry.ts'
 export type { SubagentDescendantListEntry, SubagentListEntry } from './list-children.ts'
diff --git a/packages/subagent/subagent/src/lifecycle.ts b/packages/subagent/subagent/src/lifecycle.ts
index 3482a00c0b..2e1d513d5a 100644
--- a/packages/subagent/subagent/src/lifecycle.ts
+++ b/packages/subagent/subagent/src/lifecycle.ts
@@ -18,12 +18,23 @@ import { randomUUID } from 'node:crypto'
 import type { Context } from '@deepseek-ai/cordis'
 import type { Agent } from '@deepseek-ai/dsh-agent'
 import type { ContentBlock } from '@deepseek-ai/dsh-llm'
-import { findLastMessageTurnEnd } from '@deepseek-ai/dsh-session'
+import { foldConsumedWork } from '@deepseek-ai/dsh-agent'
 import type { SessionEvent, SessionId } from '@deepseek-ai/dsh-session'
 import { finalAssistantOutput } from './assistant-output.ts'
 import { SubagentRunId } from './types.ts'
 import type { SubagentResult, SubagentRun, SubagentRunEndInfo, SubagentRunInfo } from './types.ts'
 
+/**
+ * How one Activation's residency epoch ended, as both the terminal lifecycle
+ * edge and the manager's own parent delivery report it.
+ */
+export interface ActivationTerminal {
+  /** Why this epoch's last ordinary turn ended, or `error` when teardown failed. */
+  readonly stopReason: SubagentResult['stopReason']
+  /** The epoch's final assistant content, absent when it produced none or failed. */
+  readonly output?: ContentBlock[]
+}
+
 /**
  * Lifecycle observer for one Activation's residency epoch, so continuable
  * children emit the same start/end pair as one-shot runs. Package-private: the
@@ -43,6 +54,15 @@ export interface ActivationObserver {
    * @param child - the quiescent child agent about to be released.
    */
   capture(child: Agent): void
+  /**
+   * Resolve the terminal facts {@link settle} will publish, without publishing
+   * them. The manager's parent delivery must run before the ownership release
+   * that lets the parent settle, which is earlier than the terminal edge; both
+   * therefore read one computation instead of restating the failure rule.
+   * @param failure - the teardown or durability failure, or `undefined` on success.
+   * @returns this epoch's stop reason and final assistant content.
+   */
+  terminal(failure: unknown): ActivationTerminal
   /**
    * Publish the terminal edge exactly once, pairing this epoch's {@link start},
    * after the disposal outcome is known. Called only for a resident epoch: a
@@ -165,9 +185,12 @@ export function createActivationObserver(
   let boundary = 0
   // Assigned by `capture()`, which the disposal path always runs before
   // `settle()`; a resident epoch therefore always has its facts by then.
-  let captured: { stopReason: SubagentResult['stopReason']; output?: ContentBlock[] } = {
-    stopReason: 'completed',
-  }
+  let captured: ActivationTerminal = { stopReason: 'completed' }
+  // Teardown failure overrides the epoch's own outcome and withholds its
+  // output: an answer this harness could not durably release is not a result.
+  const terminal = (failure: unknown): ActivationTerminal => failure === undefined
+    ? captured
+    : { stopReason: 'error' }
   return {
     start: (child: Agent): void => {
       boundary = child.session.events.length
@@ -181,11 +204,12 @@ export function createActivationObserver(
         ...output === undefined ? {} : { output },
       }
     },
+    terminal,
     settle: (failure: unknown): void => {
-      const output = failure === undefined ? captured.output : undefined
+      const { stopReason, output } = terminal(failure)
       emit('subagent/end', {
         ...identity,
-        stopReason: failure === undefined ? captured.stopReason : 'error',
+        stopReason,
         ...output === undefined ? {} : { lastAssistantMessage: output },
       }, parent)
     },
@@ -193,18 +217,24 @@ export function createActivationObserver(
 }
 
 /**
- * Why this child's last ordinary turn ended, for the terminal lifecycle edge.
- * The child's own `turn/end` is authoritative: teardown succeeding says nothing
- * about whether the model errored, hit its token ceiling, or was cancelled, so
- * deriving the reason from disposal would report failed work as completed.
+ * Why this child's epoch ended, for the terminal lifecycle edge and the
+ * manager's own parent delivery. The child's own log is authoritative:
+ * teardown succeeding says nothing about whether the model errored, hit its
+ * token ceiling, or was cancelled, so deriving the reason from disposal would
+ * report failed work as completed.
+ *
+ * {@link foldConsumedWork} supplies both halves the raw turn sequence cannot:
+ * which turn accounts for the work this epoch consumed, and whether accepted
+ * work was cancelled after it without any turn opening over it. A recorded
+ * failure still wins over a cancellation — stopping a child that had already
+ * failed does not turn its failure into a cancellation.
  * @param events - this epoch's own event suffix.
- * @returns its terminal stop reason; `completed` when no ordinary turn closed.
+ * @returns its terminal stop reason; `completed` only for an epoch that both
+ *   closed cleanly and had nothing left to run.
  */
 function epochStopReason(events: readonly SessionEvent[]): SubagentResult['stopReason'] {
-  const reason = findLastMessageTurnEnd(events)?.data.reason
-  // No ordinary turn closed, so nothing failed either.
-  if (reason === undefined) return 'completed'
-  switch (reason.kind) {
+  const { end, droppedUnrun } = foldConsumedWork(events)
+  switch (end?.data.reason.kind) {
     case 'max-tokens':
       return 'max-tokens'
     case 'aborted':
@@ -212,8 +242,15 @@ function epochStopReason(events: readonly SessionEvent[]): SubagentResult['stopR
       return 'aborted'
     case 'error':
       return 'error'
+    // A pre-step rejection — a hook deny, a policy plugin — discarded input
+    // this epoch had claimed: the work was declined, not done.
+    case 'blocked':
+      return 'refusal'
+    // A clean ending and no accounting turn at all share one rule: the epoch
+    // finished what it was given unless a cancelled queue says otherwise.
+    case undefined:
     case 'completed':
-      return 'completed'
+      return droppedUnrun ? 'aborted' : 'completed'
     /* v8 ignore next 3 -- `TurnEndReason` is merge-extensible, so this arm needs a
      * backend that adds a variant; treating an unnameable reason as success would
      * report failed work as completed. */
diff --git a/packages/subagent/subagent/tests/continuation.spec.ts b/packages/subagent/subagent/tests/continuation.spec.ts
index 50bd71b9cc..828717c1aa 100644
--- a/packages/subagent/subagent/tests/continuation.spec.ts
+++ b/packages/subagent/subagent/tests/continuation.spec.ts
@@ -15,7 +15,7 @@ import type { GenerateOptions, MessageId, StreamChunk } from '@deepseek-ai/dsh-l
 import { CallId, createUserMessage, LlmAdapter } from '@deepseek-ai/dsh-llm'
 import { defineTool } from '@deepseek-ai/dsh-tools'
 import InvariantService from '@deepseek-ai/dsh-invariants'
-import { MockAdapter, textResponse, toolCallResponse } from '../../../core/agent-loop/tests/mock-adapter.ts'
+import { MockAdapter, maxTokensResponse, textResponse, toolCallResponse } from '../../../core/agent-loop/tests/mock-adapter.ts'
 import SubagentService, {
   SubagentError,
   SUBAGENT_DESCRIPTOR_VERSION,
@@ -142,6 +142,18 @@ async function waitNoActivation(ctx: Context, childId: SessionId): Promise<void>
   }, { timeout: 5_000 })
 }
 
+/**
+ * Keep the top-level test parent out of a scripted model corpus. Every child
+ * settlement wakes its parent, so a suite that scripts only child responses
+ * would otherwise spend them on the parent's own turns.
+ */
+function parkParent(ctx: Context, parent: Agent): void {
+  ctx.on('agent/pre-step', async ({ agent: subject }, next) => {
+    if (subject !== parent) return next()
+    return { kind: 'reject' as const }
+  })
+}
+
 /** Observe calls at the Agent cancellation boundary without a production event. */
 function observeCancel(agent: Agent, callback: () => void): void {
   const cancel = agent.cancel.bind(agent)
@@ -531,7 +543,10 @@ describe('SubagentService.followup residency routing', () => {
     await waitNoActivation(ctx, grandchild.childId)
     await waitNoActivation(ctx, started.childId)
     const loaded = await ctx.sessionPersistence.load(started.childId)
-    expect(userTexts(loaded.events)).toEqual(['child task', 'while waiting'])
+    // This child is itself a parent, so its grandchild's settlement notice is
+    // an ordinary later user message in its log.
+    expect(userTexts(loaded.events).slice(0, 2)).toEqual(['child task', 'while waiting'])
+    expect(userTexts(loaded.events).slice(2).join('\n')).toContain('finished and will do no further work')
   })
 
   it('rejects a parent that is not the durable direct parent', async () => {
@@ -1182,6 +1197,7 @@ describe('continuable review regressions', () => {
 
   it('reports this epoch\'s own output, captured while the child was still live', async () => {
     const { ctx, parent } = await setup([textResponse('first answer'), textResponse('second answer')])
+    parkParent(ctx, parent)
     const ends: SubagentRunEndInfo[] = []
     ctx.on('subagent/end', (info) => { ends.push(info) })
 
@@ -1254,9 +1270,10 @@ describe('continuable review regressions', () => {
     await waitNoActivation(ctx, started.childId)
 
     await vi.waitFor(() => { expect(ends).toHaveLength(1) })
-    // Reading the whole session would resurrect 'first answer' here.
+    // Reading the whole session would resurrect 'first answer' here. The
+    // rejection discarded the claimed follow-up, so the epoch reads as refused.
     expect(ends[0]!.lastAssistantMessage).toBeUndefined()
-    expect(ends[0]!.stopReason).toBe('completed')
+    expect(ends[0]!.stopReason).toBe('refusal')
   })
 
   it('reports handle-disposal failure on the terminal edge', async () => {
@@ -1439,11 +1456,13 @@ describe('continuable review regressions', () => {
     await waitNoActivation(ctx, started.childId)
   })
 
-  it('reports completed when no ordinary turn closed', async () => {
+  it('reports a prompt a pre-step rejection discarded as refusal', async () => {
     const { ctx, parent } = await setup([])
+    parkParent(ctx, parent)
     const ends: SubagentRunEndInfo[] = []
     ctx.on('subagent/end', (info) => { ends.push(info) })
-    // Block admission so the child's only turn never opens.
+    // A UserPromptSubmit deny or a policy plugin: the child claims its prompt,
+    // the rejection discards it, and no step ever runs.
     ctx.on('agent/pre-step', async ({ agent: subject }, next) => {
       if (subject === parent) return next()
       return { kind: 'reject' }
@@ -1452,8 +1471,10 @@ describe('continuable review regressions', () => {
     const started = await ctx.subagents.startContinuable(startSpec(parent))
     await waitNoActivation(ctx, started.childId)
 
+    // The parent would otherwise believe a vetoed delivery was done and never
+    // resend it — the one failure the settlement promise says cannot happen.
     await vi.waitFor(() => { expect(ends).toHaveLength(1) })
-    expect(ends[0]!.stopReason).toBe('completed')
+    expect(ends[0]!.stopReason).toBe('refusal')
   })
 
   it('retains the Activation while an accepted message is still in the inbox', async () => {
@@ -1482,15 +1503,538 @@ describe('continuable review regressions', () => {
     expect(ctx.agents.get(started.childId)).toBe(child)
     releaseFirst.resolve(undefined)
     await waitNoActivation(ctx, started.childId)
-    expect(adapter.requests).toHaveLength(2)
+    // Two child turns; the third request is the parent's own turn on the
+    // settlement notice.
+    expect(adapter.requests.filter(request => request.sessionId === started.childId)).toHaveLength(2)
     const loaded = await ctx.sessionPersistence.load(started.childId)
     expect(hasUserText(loaded.events, 'queued')).toBe(true)
   })
 })
 
+/** Every settlement notice this agent received, in order, as flat text. */
+function settlementNotices(agent: Agent): { sender: string; text: string; summary: string }[] {
+  const logged = agent.session.events.flatMap(event => event.type === 'user/message' ? [event.data] : [])
+  return [...logged, ...agent.inbox.nextStep, ...agent.inbox.nextTurn].flatMap((message) => {
+    if (message.source.kind !== 'subagent-settled') return []
+    return [{
+      sender: message.source.senderSessionId,
+      summary: message.source.summary,
+      text: message.content.flatMap(block => block.type === 'text' ? [block.text] : []).join('\n'),
+    }]
+  })
+}
+
+describe('continuable settlement delivery', () => {
+  it('tells the parent what the child finished with, without being asked', async () => {
+    const { ctx, parent } = await setup([textResponse('the answer'), textResponse('parent ack')])
+    const started = await ctx.subagents.startContinuable(startSpec(parent))
+    await waitNoActivation(ctx, started.childId)
+
+    await vi.waitFor(() => { expect(settlementNotices(parent)).toHaveLength(1) })
+    const notice = settlementNotices(parent)[0]!
+    expect(notice.sender).toBe(started.childId)
+    expect(notice.text).toBe(
+      `Background subagent ${started.childId} finished and will do no further work unless you send it more.`
+      + '\nIts closing message:\nthe answer',
+    )
+    // The collapsed row states the outcome without the child's content.
+    expect(notice.summary).toBe(
+      `Background subagent ${started.childId} finished and will do no further work unless you send it more.`,
+    )
+  })
+
+  it('delivers even when the child already reported for itself', async () => {
+    const { ctx, parent } = await setup([textResponse('the answer'), textResponse('parent ack')])
+    const started = await ctx.subagents.startContinuable(startSpec(parent))
+    const child = await vi.waitFor(() => {
+      const live = ctx.agents.get(started.childId)
+      expect(live).toBeDefined()
+      return live!
+    })
+    await ctx.subagents.reportFrom(child, message('an explicit report'), {
+      delivery: 'quiet',
+      signal: testSignal,
+    })
+    await waitNoActivation(ctx, started.childId)
+
+    // The contract is unconditional precisely so the parent-side tool
+    // description can promise it; bookkeeping "did it report?" would make the
+    // promise conditional on a channel this manager does not own.
+    await vi.waitFor(() => { expect(settlementNotices(parent)).toHaveLength(1) })
+  })
+
+  it('delivers the terminal reason when the child never had a chance to report', async () => {
+    const { ctx, parent } = await setup([maxTokensResponse('half an ans'), textResponse('parent ack')])
+    const started = await ctx.subagents.startContinuable(startSpec(parent))
+    await waitNoActivation(ctx, started.childId)
+
+    await vi.waitFor(() => { expect(settlementNotices(parent)).toHaveLength(1) })
+    expect(settlementNotices(parent)[0]!.text).toBe(
+      `Background subagent ${started.childId} ran out of room before it finished.`
+      + '\nIts closing message:\nhalf an ans',
+    )
+  })
+
+  it('tells the parent a policy-rejected delivery was declined, not finished', async () => {
+    const { ctx, parent } = await setup([textResponse('parent ack')])
+    // A pre-step rejection on the child — a UserPromptSubmit deny, a policy
+    // plugin — discards the claimed prompt without running it.
+    ctx.on('agent/pre-step', async ({ agent: subject }, next) => {
+      if (subject === parent) return next()
+      return { kind: 'reject' }
+    })
+
+    const started = await ctx.subagents.startContinuable(startSpec(parent))
+    await waitNoActivation(ctx, started.childId)
+
+    await vi.waitFor(() => { expect(settlementNotices(parent)).toHaveLength(1) })
+    expect(settlementNotices(parent)[0]!.text).toBe(
+      `Background subagent ${started.childId} declined the task.`
+      + '\nIt left no closing message.',
+    )
+  })
+
+  it('reports a turn that failed before reaching its first step', async () => {
+    const releaseFirst = Promise.withResolvers<undefined>()
+    const adapter = new GatedAdapter([
+      { chunks: textResponse('the answer'), gate: releaseFirst.promise },
+      { chunks: textResponse('parent ack') },
+    ])
+    const { ctx, parent } = await setupWith(adapter)
+    // The shipped durability checkpoint (`dsh-session-checkpoint-policy`) is
+    // fail-closed at the step boundary, so a rejected write ends the turn after
+    // it claimed its messages and before it entered a step.
+    ctx.on('agent/pre-step', async ({ agent: subject, turn }, next) => {
+      if (subject.session.header.parentSession === undefined || turn < 2) return next()
+      throw new Error('ENOSPC: no space left on device')
+    })
+
+    const started = await ctx.subagents.startContinuable(startSpec(parent))
+    await followup(ctx, parent, started.childId, message('second task'))
+    releaseFirst.resolve(undefined)
+    await waitNoActivation(ctx, started.childId)
+
+    await vi.waitFor(() => { expect(settlementNotices(parent)).toHaveLength(1) })
+    // The parent must not be told the child finished: the delivery it is still
+    // waiting on was claimed out of the inbox and then swallowed by the failure.
+    const child = await ctx.sessionPersistence.load(started.childId)
+    expect(hasUserText(child.events, 'second task')).toBe(false)
+    expect(settlementNotices(parent)[0]!.text).toBe(
+      `Background subagent ${started.childId} failed before it finished.`
+      + '\nIts closing message:\nthe answer',
+    )
+  })
+
+  it('reports accepted work cut short before its first step as stopped', async () => {
+    const releaseFirst = Promise.withResolvers<undefined>()
+    const releaseCheckpoint = Promise.withResolvers<undefined>()
+    const atCheckpoint = Promise.withResolvers<undefined>()
+    const adapter = new GatedAdapter([{ chunks: textResponse('the answer'), gate: releaseFirst.promise }])
+    const { ctx, parent } = await setupWith(adapter)
+    // A step-boundary participant — the shipped durability checkpoint, a hook,
+    // prompt assembly — holding the child's second turn open before its first
+    // step, which is where teardown cancellation then catches it.
+    ctx.on('agent/pre-step', async ({ agent: subject, turn }, next) => {
+      if (subject.session.header.parentSession === undefined || turn < 2) return next()
+      atCheckpoint.resolve(undefined)
+      await releaseCheckpoint.promise
+      return next()
+    })
+
+    const started = await ctx.subagents.startContinuable(startSpec(parent))
+    // Queued while turn 1 still runs, so turn 2 opens and claims it without a
+    // second model call: the Activation is mid-turn when the drain cancels it.
+    await followup(ctx, parent, started.childId, message('second task'))
+    releaseFirst.resolve(undefined)
+    await atCheckpoint.promise
+    const drained = drainManager(ctx)
+    releaseCheckpoint.resolve(undefined)
+    await drained
+
+    // Turn 2 leaves a balanced no-step `aborted` end, so the log alone would
+    // answer with turn 1's clean completion and tell the parent its still-unrun
+    // task had finished.
+    await vi.waitFor(() => { expect(settlementNotices(parent)).toHaveLength(1) })
+    expect(settlementNotices(parent)[0]!.text).toBe(
+      `Background subagent ${started.childId} was stopped before it finished.`
+      + '\nIts closing message:\nthe answer',
+    )
+  })
+
+  it('reports a child stopped before it ever reached the model as stopped', async () => {
+    const releaseCheckpoint = Promise.withResolvers<undefined>()
+    const atCheckpoint = Promise.withResolvers<undefined>()
+    const { ctx, parent } = await setupWith(new GatedAdapter([]))
+    ctx.on('agent/pre-step', async ({ agent: subject }, next) => {
+      if (subject.session.header.parentSession === undefined) return next()
+      atCheckpoint.resolve(undefined)
+      await releaseCheckpoint.promise
+      return next()
+    })
+
+    const started = await ctx.subagents.startContinuable(startSpec(parent))
+    await atCheckpoint.promise
+    const drained = drainManager(ctx)
+    releaseCheckpoint.resolve(undefined)
+    await drained
+
+    // This epoch closed no stepped turn at all, which on its own reads as "had
+    // nothing to report"; only the interruption distinguishes it from a child
+    // that genuinely finished with no output.
+    await vi.waitFor(() => { expect(settlementNotices(parent)).toHaveLength(1) })
+    expect(settlementNotices(parent)[0]!.text).toBe(
+      `Background subagent ${started.childId} was stopped before it finished.`
+      + '\nIt left no closing message.',
+    )
+  })
+
+  it('reports a child an ancestor interrupted before its first step as stopped', async () => {
+    const atCheckpoint = Promise.withResolvers<undefined>()
+    const releaseCheckpoint = Promise.withResolvers<undefined>()
+    const adapter = new GatedAdapter([{ chunks: textResponse('parent ack') }])
+    const { ctx, parent } = await setupWith(adapter)
+    ctx.on('agent/pre-step', async ({ agent: subject }, next) => {
+      if (subject.session.header.parentSession === undefined) return next()
+      atCheckpoint.resolve(undefined)
+      await releaseCheckpoint.promise
+      return next()
+    })
+
+    const started = await ctx.subagents.startContinuable(startSpec(parent))
+    await atCheckpoint.promise
+    // The shipped interrupt path: nothing about it runs inside this manager, so
+    // no pre-cancel sample could see it — the child's own log has to say so.
+    ctx.subagents.interrupt(started.childId, { kind: 'ancestor', agent: parent })
+    releaseCheckpoint.resolve(undefined)
+    await waitNoActivation(ctx, started.childId)
+
+    await vi.waitFor(() => { expect(settlementNotices(parent)).toHaveLength(1) })
+    expect(settlementNotices(parent)[0]!.text).toBe(
+      `Background subagent ${started.childId} was stopped before it finished.`
+      + '\nIt left no closing message.',
+    )
+  })
+
+  it('reports accepted work cancelled before any turn could open as stopped', async () => {
+    const releaseChild = Promise.withResolvers<undefined>()
+    const releaseGrandchild = Promise.withResolvers<undefined>()
+    const releaseMaintenance = Promise.withResolvers<undefined>()
+    const adapter = new GatedAdapter([
+      { chunks: textResponse('the answer'), gate: releaseChild.promise },
+      { chunks: textResponse('grandchild'), gate: releaseGrandchild.promise },
+    ])
+    const { ctx, parent } = await setupWith(adapter)
+    const started = await ctx.subagents.startContinuable(startSpec(parent))
+    const child = await vi.waitFor(() => {
+      const live = ctx.agents.get(started.childId)
+      expect(live).toBeDefined()
+      return live!
+    })
+    // A descendant keeps the child resident once its own turn closes, so the
+    // maintenance phase below is reachable without racing settlement.
+    const grandchild = await ctx.subagents.startContinuable(startSpec(child))
+    await vi.waitFor(() => { expect(ctx.agents.get(grandchild.childId)).toBeDefined() })
+    releaseChild.resolve(undefined)
+    await vi.waitFor(() => { expect(child.status).toBe('idle') })
+
+    // Context maintenance folds into `idle` and defers waking work, so this
+    // delivery is accepted with no turn to claim it.
+    const maintaining = child.runMaintenance(async () => { await releaseMaintenance.promise })
+    await followup(ctx, parent, started.childId, message('never runs'))
+    const drained = drainManager(ctx)
+    releaseMaintenance.resolve(undefined)
+    releaseGrandchild.resolve(undefined)
+    await maintaining
+    await drained
+
+    // Turn 1 closed cleanly and no later turn opened, so the cancelled queue is
+    // the only record that this epoch was cut short.
+    expect(hasUserText(child.session.events, 'never runs')).toBe(false)
+    await vi.waitFor(() => { expect(settlementNotices(parent)).toHaveLength(1) })
+    expect(settlementNotices(parent)[0]!.text).toBe(
+      `Background subagent ${started.childId} was stopped before it finished.`
+      + '\nIts closing message:\nthe answer',
+    )
+  })
+
+  it('withholds an outcome the harness could not durably release', async () => {
+    const { ctx, parent } = await setup([textResponse('the answer'), textResponse('parent ack')])
+    const started = await ctx.subagents.startContinuable(startSpec(parent))
+    const manager = (ctx.subagents as unknown as {
+      continuations: { activations: Map<SessionId, { handle: { dispose(): Promise<void> } }> }
+    }).continuations
+    const activation = await vi.waitFor(() => {
+      const live = manager.activations.get(started.childId)
+      expect(live).toBeDefined()
+      return live!
+    })
+    const dispose = activation.handle.dispose.bind(activation.handle)
+    activation.handle.dispose = async () => {
+      await dispose()
+      throw new Error('scope unwind failed')
+    }
+
+    await waitNoActivation(ctx, started.childId)
+    await vi.waitFor(() => { expect(settlementNotices(parent)).toHaveLength(1) })
+    expect(settlementNotices(parent)[0]!.text).toBe(
+      `Background subagent ${started.childId} failed before it finished.\nIt left no closing message.`,
+    )
+  })
+
+  it('gives an idle parent one ordinary turn on the notice', async () => {
+    const { ctx, parent, adapter } = await setup([textResponse('the answer'), textResponse('parent ack')])
+    const turnStarts: number[] = []
+    ctx.on('session/event', (session, event) => {
+      if (session.id === parent.id && event.type === 'turn/start') turnStarts.push(event.data.turn)
+    })
+
+    const started = await ctx.subagents.startContinuable(startSpec(parent))
+    await waitNoActivation(ctx, started.childId)
+    await vi.waitFor(() => {
+      expect(adapter.requests.filter(request => request.sessionId === parent.id)).toHaveLength(1)
+    })
+    expect(turnStarts).toEqual([1])
+  })
+
+  it('batches simultaneous notices into one step of a busy parent', async () => {
+    const releaseChildren = Promise.withResolvers<undefined>()
+    const releaseParent = Promise.withResolvers<undefined>()
+    const adapter = new GatedAdapter([
+      { chunks: textResponse('parent works'), gate: releaseParent.promise },
+      { chunks: textResponse('first child'), gate: releaseChildren.promise },
+      { chunks: textResponse('second child'), gate: releaseChildren.promise },
+      { chunks: textResponse('parent reacts') },
+    ])
+    const { ctx, parent } = await setupWith(adapter)
+    // Open a parent turn first, so both notices arrive while it is running.
+    parent.followup(createUserMessage({ content: message('start working'), source: { kind: 'user' } }))
+    await vi.waitFor(() => { expect(parent.status).toBe('running') })
+
+    const first = await ctx.subagents.startContinuable(startSpec(parent))
+    const second = await ctx.subagents.startContinuable(startSpec(parent))
+    releaseChildren.resolve(undefined)
+    await waitNoActivation(ctx, first.childId)
+    await waitNoActivation(ctx, second.childId)
+
+    // Both notices are waiting for the same step boundary, not two turns.
+    expect(parent.inbox.nextStep).toHaveLength(2)
+    expect(parent.inbox.nextTurn).toHaveLength(0)
+    const turnStarts: number[] = []
+    ctx.on('session/event', (session, event) => {
+      if (session.id === parent.id && event.type === 'turn/start') turnStarts.push(event.data.turn)
+    })
+    releaseParent.resolve(undefined)
+    await vi.waitFor(() => { expect(settlementNotices(parent)).toHaveLength(2) })
+    expect(turnStarts).toEqual([])
+    // Both children released together, so which settles first is not ordered.
+    expect(new Set(settlementNotices(parent).map(entry => entry.sender)))
+      .toEqual(new Set([first.childId, second.childId]))
+  })
+
+  it('holds a maintaining parent live until it can read the notice', async () => {
+    const releaseFirst = Promise.withResolvers<undefined>()
+    const releaseSecond = Promise.withResolvers<undefined>()
+    const adapter = new GatedAdapter([
+      { chunks: textResponse('outer') },
+      { chunks: textResponse('first inner'), gate: releaseFirst.promise },
+      { chunks: textResponse('second inner'), gate: releaseSecond.promise },
+      { chunks: textResponse('outer reacts') },
+      { chunks: textResponse('root reacts') },
+    ])
+    const { ctx, parent } = await setupWith(adapter)
+    const outer = await ctx.subagents.startContinuable(startSpec(parent))
+    const middle = await vi.waitFor(() => {
+      const live = ctx.agents.get(outer.childId)
+      expect(live).toBeDefined()
+      return live!
+    })
+    const first = await ctx.subagents.startContinuable(startSpec(middle))
+    const second = await ctx.subagents.startContinuable(startSpec(middle))
+    await vi.waitFor(() => { expect(middle.status).toBe('idle') })
+
+    // `Agent.status` folds maintenance into `idle`, and a waking send behind it
+    // only arms a deferred wake. The first child's release moves the middle
+    // Activation's settlement watcher onto its quiescence race; the second one
+    // then arrives at exactly the point where an unaccounted delivery would be
+    // judged quiet, settled, and cancelled — clearing the inbox it sits in.
+    const maintaining = Promise.withResolvers<undefined>()
+    const maintenance = middle.runMaintenance(async () => { await maintaining.promise })
+    releaseFirst.resolve(undefined)
+    await waitNoActivation(ctx, first.childId)
+    releaseSecond.resolve(undefined)
+    await waitNoActivation(ctx, second.childId)
+    expect(ctx.agents.get(outer.childId)).toBe(middle)
+
+    maintaining.resolve(undefined)
+    await maintenance
+    await vi.waitFor(() => { expect(settlementNotices(middle)).toHaveLength(2) })
+    expect(settlementNotices(middle).map(entry => entry.sender))
+      .toEqual([first.childId, second.childId])
+    await waitNoActivation(ctx, outer.childId)
+  })
+
+  it('delivers before releasing the ownership that lets the parent settle', async () => {
+    const releaseChild = Promise.withResolvers<undefined>()
+    const adapter = new GatedAdapter([
+      { chunks: textResponse('outer') },
+      { chunks: textResponse('inner'), gate: releaseChild.promise },
+      { chunks: textResponse('outer reacts') },
+    ])
+    const { ctx, parent } = await setupWith(adapter)
+    const outer = await ctx.subagents.startContinuable(startSpec(parent))
+    const middle = await vi.waitFor(() => {
+      const live = ctx.agents.get(outer.childId)
+      expect(live).toBeDefined()
+      return live!
+    })
+    const inner = await ctx.subagents.startContinuable(startSpec(middle))
+    await vi.waitFor(() => { expect(middle.status).toBe('idle') })
+
+    const manager = (ctx.subagents as unknown as {
+      continuations: { activations: Map<SessionId, { ownedChildren: Set<SessionId> }> }
+    }).continuations
+    let ownedAtDelivery: SessionId[] | undefined
+    ctx.on('agent/inbox/inserted', ({ agent, message }) => {
+      if (agent !== middle || message.source.kind !== 'subagent-settled') return
+      ownedAtDelivery = [...manager.activations.get(middle.id)!.ownedChildren]
+    })
+
+    releaseChild.resolve(undefined)
+    await waitNoActivation(ctx, inner.childId)
+    // Still owned at delivery: the parent is structurally unable to settle in
+    // the window the notice crosses, rather than winning a race against it.
+    expect(ownedAtDelivery).toEqual([inner.childId])
+    await waitNoActivation(ctx, outer.childId)
+  })
+
+  it('does not wake a parent whose own teardown already began', async () => {
+    const hold = Promise.withResolvers<undefined>()
+    const adapter = new GatedAdapter([{ chunks: textResponse('interrupted'), gate: hold.promise }])
+    const { ctx, parent } = await setupWith(adapter)
+    const started = await ctx.subagents.startContinuable(startSpec(parent))
+    await vi.waitFor(() => { expect(ctx.agents.get(started.childId)).toBeDefined() })
+
+    const drained = drainManager(ctx)
+    hold.resolve(undefined)
+    await drained
+
+    // Delivered and durably logged, but no turn: waking a parent the host is
+    // about to dispose spends a model request nothing reads. What happens to the
+    // message when that parent is disposed next is pinned by the test below.
+    expect(settlementNotices(parent)).toHaveLength(1)
+    expect(settlementNotices(parent)[0]!.text).toBe(
+      `Background subagent ${started.childId} was stopped before it finished.`
+      + '\nIt left no closing message.',
+    )
+    expect(parent.session.events.some(event => event.type === 'agent/inbox/spliced')).toBe(true)
+    expect(parent.session.events.some(event => event.type === 'turn/start')).toBe(false)
+    expect(parent.status).toBe('idle')
+  })
+
+  it('does not wake a parent below a scoped teardown root', async () => {
+    const hold = Promise.withResolvers<undefined>()
+    const adapter = new GatedAdapter([{ chunks: textResponse('interrupted'), gate: hold.promise }])
+    const { ctx, parent } = await setupWith(adapter)
+    const started = await ctx.subagents.startContinuable(startSpec(parent))
+    await vi.waitFor(() => { expect(ctx.agents.get(started.childId)).toBeDefined() })
+
+    const drained = ctx.subagents.drainContinuableDescendants([parent])
+    hold.resolve(undefined)
+    await drained
+
+    expect(settlementNotices(parent)).toHaveLength(1)
+    expect(parent.session.events.some(event => event.type === 'turn/start')).toBe(false)
+  })
+
+  it('records but cannot deliver a teardown notice once the parent is disposed too', async () => {
+    const hold = Promise.withResolvers<undefined>()
+    const adapter = new GatedAdapter([{ chunks: textResponse('interrupted'), gate: hold.promise }])
+    const { ctx } = await setupWith(adapter)
+    const parentId = SessionId('closing-parent')
+    const host = await ctx.agents.create({
+      sessionId: parentId,
+      agentOptions: { provider: 'mock', model: 'mock' },
+    })
+    const started = await ctx.subagents.startContinuable(startSpec(host.agent))
+    await vi.waitFor(() => { expect(ctx.agents.get(started.childId)).toBeDefined() })
+
+    const drained = ctx.subagents.drainContinuableDescendants([host.agent])
+    hold.resolve(undefined)
+    await drained
+    expect(settlementNotices(host.agent)).toHaveLength(1)
+
+    // Disposal is a `keepInbox: false` cancel, so it durably cancels the notice
+    // it never claimed. Teardown delivery therefore reaches a parent that is
+    // still resident — a resumed one reads the log, not a pending message — and
+    // no wording anywhere may promise otherwise.
+    await host.dispose()
+    const resumed = await ctx.agents.resume({
+      resumeSessionId: parentId,
+      agentOptions: { provider: 'mock', model: 'mock' },
+    })
+    expect(settlementNotices(resumed.agent)).toEqual([])
+    await resumed.dispose()
+    // The account is still in the durable log: delivered, then cancelled unread.
+    const persisted = await ctx.sessionPersistence.load(parentId)
+    expect(persisted.events.flatMap(event => event.type === 'agent/inbox/spliced'
+      ? [{ inserted: event.data.inserted.length, removed: event.data.removedCount ?? 0 }]
+      : [])).toEqual([{ inserted: 1, removed: 0 }, { inserted: 0, removed: 1 }])
+  })
+
+  it('drops the notice without disturbing teardown when the parent is gone', async () => {
+    const releaseChild = Promise.withResolvers<undefined>()
+    const adapter = new GatedAdapter([{ chunks: textResponse('answer'), gate: releaseChild.promise }])
+    const { ctx } = await setupWith(adapter)
+    const warnings: string[] = []
+    ctx.logger.warn = (text: string) => { warnings.push(text) }
+    const host = await ctx.agents.create({
+      sessionId: SessionId('disposable-parent'),
+      agentOptions: { provider: 'mock', model: 'mock' },
+    })
+    const started = await ctx.subagents.startContinuable(startSpec(host.agent))
+    const ends: SubagentRunEndInfo[] = []
+    ctx.on('subagent/end', (info) => { ends.push(info) })
+
+    releaseChild.resolve(undefined)
+    await host.dispose()
+    await waitNoActivation(ctx, started.childId)
+    await vi.waitFor(() => { expect(ends).toHaveLength(1) })
+    expect(warnings).toEqual([])
+  })
+
+  it('logs a rejected notice instead of failing the child\'s teardown', async () => {
+    const { ctx, parent } = await setup([textResponse('the answer')])
+    const warnings: string[] = []
+    ctx.logger.warn = (text: string) => { warnings.push(text) }
+    vi.spyOn(parent, 'followup').mockImplementation(() => {
+      throw new Error('parent closed during delivery')
+    })
+    const ends: SubagentRunEndInfo[] = []
+    ctx.on('subagent/end', (info) => { ends.push(info) })
+
+    const started = await ctx.subagents.startContinuable(startSpec(parent))
+    await waitNoActivation(ctx, started.childId)
+    await vi.waitFor(() => { expect(ends).toHaveLength(1) })
+    expect(ends[0]!.stopReason).toBe('completed')
+    expect(warnings.some(warning => warning.includes('settlement notice was not delivered'))).toBe(true)
+  })
+
+  it('stays silent about a child the caller was told does not exist', async () => {
+    const { ctx, parent } = await setup([])
+    const drains: Promise<void>[] = []
+    ctx.on('subagent/start', () => { drains.push(drainManager(ctx)) })
+
+    await expect(ctx.subagents.startContinuable(startSpec(parent)))
+      .rejects.toMatchObject({ code: 'DRAINING' })
+    await Promise.all(drains)
+    expect(settlementNotices(parent)).toEqual([])
+  })
+})
+
 describe('continuable lifecycle observation', () => {
   it('emits one paired start/end per residency epoch', async () => {
     const { ctx, parent } = await setup([textResponse('first'), textResponse('second')])
+    parkParent(ctx, parent)
     const starts: SubagentRunInfo[] = []
     const ends: SubagentRunEndInfo[] = []
     ctx.on('subagent/start', (info) => { starts.push(info) })
@@ -1510,6 +2054,8 @@ describe('continuable lifecycle observation', () => {
     expect(starts.map(info => info.provider)).toEqual(['spawn', 'spawn'])
     // Each end pairs its own start's runId.
     expect(ends.map(info => info.runId)).toEqual(starts.map(info => info.runId))
+    // Both epochs ran their own scripted response; neither exhausted the corpus.
+    expect(ends.map(info => info.stopReason)).toEqual(['completed', 'completed'])
   })
 })
 
diff --git a/packages/subagent/tool-subagent-control/README.i18n.yaml b/packages/subagent/tool-subagent-control/README.i18n.yaml
index 885345b576..fbdd5b78db 100644
--- a/packages/subagent/tool-subagent-control/README.i18n.yaml
+++ b/packages/subagent/tool-subagent-control/README.i18n.yaml
@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write packages/subagent/tool-subagent-control/README.md
-README.md: 460617378841b791f7d8750962a9e4549643928e
-README.zh.md: 5939a1636a5f92dc797e5788c88bba7048ec7c96
+README.md: 91f8d23ac092049e5315418070cdaae025054860
+README.zh.md: 981e3f4efa2a904918d3f4174a2188fc2965c626
diff --git a/packages/subagent/tool-subagent-control/README.md b/packages/subagent/tool-subagent-control/README.md
index 4606173788..91f8d23ac0 100644
--- a/packages/subagent/tool-subagent-control/README.md
+++ b/packages/subagent/tool-subagent-control/README.md
@@ -8,7 +8,7 @@ The tool performs no lifecycle routing — residency and cold resume belong to t
 
 `interrupt_agent(agent_id)` passes `exec.agent` as the exact live ancestor authority for `ctx.subagents.interrupt()`: the target may be a direct child or a deeper descendant, and the service — never this tool — verifies the caller against the target Activation's recorded lineage. Only the target's current turn stops (`keepInbox`): queued messages stay parked until a later `send_message`, published descendants keep running, and the child stays available for follow-ups. The call returns as soon as the stop request is accepted, without waiting for target quiescence; an absent or already-settled target is an accepted no-op, while self, sibling, stale, and non-ancestor callers become errored results.
 
-`list_agents` takes one optional `scope` argument, derives the root id from the calling agent, and projects the service catalog to continuable children without a cursor. The default `children` scope reads `ctx.subagents.listChildren()`; `descendants` reads `ctx.subagents.listDescendants()`, whose one-corpus walk crosses ordinary sessions and one-shot children and renders surviving rows in stable pre-order with `parent=<id> depth=<n>`. The `parent` annotation is the durable direct-parent session id and may name an ordinary session omitted from the output. For the calling agent, only depth-1 child entries are `send_message` candidates; deeper child entries are `interrupt_agent` candidates only. Status comes from the live Agent registry: `running` (active driver), `idle` (resident between turns, possibly waiting on agents it started), `complete` (storage only). The service result also contains one-shot session-backed subagents for consumers such as a UI, but those entries are omitted from this model tool because they cannot accept `send_message`. Diagnostics remain visible, with positions in the descendants scope. Durable identity and mode come from each child's descriptor, while delivery-time authority and Activation ownership checks remain the service's.
+`list_agents` takes one optional `scope` argument, derives the root id from the calling agent, and projects the service catalog to continuable children without a cursor. The default `children` scope reads `ctx.subagents.listChildren()`; `descendants` reads `ctx.subagents.listDescendants()`, whose one-corpus walk crosses ordinary sessions and one-shot children and renders surviving rows in stable pre-order with `parent=<id> depth=<n>`. The `parent` annotation is the durable direct-parent session id and may name an ordinary session omitted from the output. For the calling agent, only depth-1 child entries are `send_message` candidates; deeper child entries are `interrupt_agent` candidates only. Status comes from the live Agent registry: `running` (active driver), `idle` (resident between turns, possibly waiting on agents it started), or `ready` (storage only and resumable rather than terminal). The service result also contains one-shot session-backed subagents for consumers such as a UI, but those entries are omitted from this model tool because they cannot accept `send_message`. Diagnostics remain visible, with positions in the descendants scope. Durable identity and mode come from each child's descriptor, while delivery-time authority and Activation ownership checks remain the service's.
 
 ## Model Experience
 
@@ -58,7 +58,7 @@ Append-only; newly visible content follows the reusable request prefix and does
 
 #### What the model sees
 
-One line per continuable child in stable catalog order: `<id> [<status>] — <label>` (`running` = active driver, `idle` = resident between turns, `complete` = storage only; a direct child in that state can be resumed by `send_message`), plus `<id> [diagnostic: <reason>]` for a candidate that could not be read (`corrupt`, `unsupported`, or `unavailable`). The `descendants` scope inserts ` parent=<id> depth=<n>` before the label dash on every line, in pre-order. One-shot children are intentionally absent; `(no subagents)` means no continuable child or diagnostic survived the projection. Diagnostics never expose descriptor contents.
+One line per continuable child in stable catalog order: `<id> [<status>] — <label>` (`running` = active driver, `idle` = resident between turns, `ready` = storage only; resumable rather than terminal, not a result waiting to be collected — a direct child in that state can be resumed by `send_message`), plus `<id> [diagnostic: <reason>]` for a candidate that could not be read (`corrupt`, `unsupported`, or `unavailable`). The `descendants` scope inserts ` parent=<id> depth=<n>` before the label dash on every line, in pre-order. One-shot children are intentionally absent; `(no subagents)` means no continuable child or diagnostic survived the projection. Diagnostics never expose descriptor contents.
 
 #### Token effect
 
@@ -72,5 +72,5 @@ Append-only; each result follows the reusable request prefix.
 
 - **A queued message has no independent result** — acceptance returns only its inbox `messageId`; the child's work lands in the durable child Session and is never collected through this tool. A child granted `report` may send selected content back separately, but that message is not this call's result.
 - **No steering of the current turn** — every message opens a later FIFO turn, so a message sent while the child is working runs only after its current turn finishes and cannot redirect it.
-- **Listing is a snapshot, not a delivery promise** — it may race publication, disposal, or a later message, and another process may activate a child this process reports as `complete`; cross-process accuracy requires a shared lease. `interrupt_agent` performs the authoritative live-lineage check itself, so discovery staleness cannot grant authority.
+- **Listing is a snapshot, not a delivery promise** — it may race publication, disposal, or a later message, and another process may activate a child this process reports as `ready`; cross-process accuracy requires a shared lease. `interrupt_agent` performs the authoritative live-lineage check itself, so discovery staleness cannot grant authority.
 - **No pagination or deletion** — the complete stably ordered set is returned, and persisted children remain listed for as long as their sessions remain in persistence; a service-level bound or delete operation is a later product decision.
diff --git a/packages/subagent/tool-subagent-control/README.zh.md b/packages/subagent/tool-subagent-control/README.zh.md
index 5939a1636a..981e3f4efa 100644
--- a/packages/subagent/tool-subagent-control/README.zh.md
+++ b/packages/subagent/tool-subagent-control/README.zh.md
@@ -8,7 +8,7 @@
 
 `interrupt_agent(agent_id)` 将 `exec.agent` 作为 `ctx.subagents.interrupt()` 的确切在线 ancestor 授权传入：目标可以是直接 child 或更深的后代，由服务——而不是本工具——依据目标 Activation 记录的 lineage 校验调用方。只有目标的当前轮次会停止（`keepInbox`）：已排队的消息保持暂停直到之后的 `send_message`，已发布的后代继续运行，child 也仍可接受后续消息。调用在停止请求被接受后立即返回，不等待目标完全停稳；目标不存在或已结算是被接受的 no-op，而 self、sibling、过期与非 ancestor 调用方会成为出错结果。
 
-`list_agents` 接受一个可选的 `scope` 参数，会从调用它的 agent 推导根 id，并且不使用 cursor，将服务目录投影为可继续 child。默认的 `children` scope 读取 `ctx.subagents.listChildren()`；`descendants` 读取 `ctx.subagents.listDescendants()`，其单份语料的遍历会穿过普通会话与一次性 child，并按稳定 pre-order 以 `parent=<id> depth=<n>` 渲染保留下来的条目。`parent` 注释是持久化直接 parent 会话 id，可能指向输出中省略的普通会话。对于调用本工具的 agent，只有 depth-1 child 条目可作为 `send_message` 候选；更深的 child 条目只能作为 `interrupt_agent` 候选。状态来自在线 Agent 注册表：`running`（driver 活跃）、`idle`（驻留但处于轮次之间，可能在等待它启动的 agent）、`complete`（仅存于存储）。服务结果还包含由会话支撑的一次性 subagent，以供 UI 等消费方使用；但这些条目无法接受 `send_message`，因此会从这个模型工具中排除。diagnostic 仍然可见，并在 descendants scope 中带有位置。持久化身份和模式来自每个子 agent 的描述符，消息送达时的鉴权和 Activation 所有权检查仍归服务负责。
+`list_agents` 接受一个可选的 `scope` 参数，会从调用它的 agent 推导根 id，并且不使用 cursor，将服务目录投影为可继续 child。默认的 `children` scope 读取 `ctx.subagents.listChildren()`；`descendants` 读取 `ctx.subagents.listDescendants()`，其单份语料的遍历会穿过普通会话与一次性 child，并按稳定 pre-order 以 `parent=<id> depth=<n>` 渲染保留下来的条目。`parent` 注释是持久化直接 parent 会话 id，可能指向输出中省略的普通会话。对于调用本工具的 agent，只有 depth-1 child 条目可作为 `send_message` 候选；更深的 child 条目只能作为 `interrupt_agent` 候选。状态来自在线 Agent 注册表：`running`（driver 活跃）、`idle`（驻留但处于轮次之间，可能在等待它启动的 agent）或 `ready`（仅存于存储，表示可恢复而非终态）。服务结果还包含由会话支撑的一次性 subagent，以供 UI 等消费方使用；但这些条目无法接受 `send_message`，因此会从这个模型工具中排除。diagnostic 仍然可见，并在 descendants scope 中带有位置。持久化身份和模式来自每个子 agent 的描述符，消息送达时的鉴权和 Activation 所有权检查仍归服务负责。
 
 ## 模型体验
 
@@ -58,7 +58,7 @@
 
 #### 模型看到的内容
 
-按稳定目录顺序，每个可继续 child 占一行：渲染为 `<id> [<status>] — <label>`（`running` 表示 driver 活跃，`idle` 表示驻留但处于轮次之间，`complete` 表示仅存于存储；处于该状态的直接 child 可通过 `send_message` 恢复），另为无法读取的候选项渲染 `<id> [diagnostic: <reason>]`（`corrupt`、`unsupported` 或 `unavailable`）。`descendants` scope 会在每行 label 破折号之前插入 ` parent=<id> depth=<n>`，按 pre-order 排列。一次性 child 会被有意排除；`(no subagents)` 表示投影后没有留下可继续 child 或 diagnostic。诊断信息绝不会暴露描述符内容。
+按稳定目录顺序，每个可继续 child 占一行：渲染为 `<id> [<status>] — <label>`（`running` 表示 driver 活跃，`idle` 表示驻留但处于轮次之间，`ready` 表示仅存于存储；可恢复而非终态，也不表示有结果等待收集——处于该状态的直接 child 可通过 `send_message` 恢复），另为无法读取的候选项渲染 `<id> [diagnostic: <reason>]`（`corrupt`、`unsupported` 或 `unavailable`）。`descendants` scope 会在每行 label 破折号之前插入 ` parent=<id> depth=<n>`，按 pre-order 排列。一次性 child 会被有意排除；`(no subagents)` 表示投影后没有留下可继续 child 或 diagnostic。诊断信息绝不会暴露描述符内容。
 
 #### Token 影响
 
@@ -72,5 +72,5 @@
 
 - **已排队的消息没有独立结果**：接受时只返回其 inbox `messageId`；子 agent 的工作会落入持久化子 agent 会话，绝不会通过本工具收集。获得 `report` 的子 agent 可以单独发回选定内容，但该消息不是本次调用的结果。
 - **不对当前轮次进行 steering（中途引导）**：每条消息都会开启后续 FIFO 轮次，因此在子 agent 工作时发送的消息只会在其当前轮次结束后运行，无法将其重定向。
-- **列表是快照，而非投递承诺**：它可能与发布、dispose（资源释放）或后续消息发生竞态，另一个进程也可能激活当前进程报告为 `complete` 的 child；跨进程准确性需要共享租约。`interrupt_agent` 自己执行权威的在线 lineage 检查，因此过期的发现结果不会授予权限。
+- **列表是快照，而非投递承诺**：它可能与发布、dispose（资源释放）或后续消息发生竞态，另一个进程也可能激活当前进程报告为 `ready` 的 child；跨进程准确性需要共享租约。`interrupt_agent` 自己执行权威的在线 lineage 检查，因此过期的发现结果不会授予权限。
 - **没有分页或删除**：系统返回完整且稳定排序的集合；只要 child 会话仍在持久化存储中，它就会继续出现在列表中，服务级上限或删除操作留待后续产品决策。
diff --git a/packages/subagent/tool-subagent-control/src/list-agents.ts b/packages/subagent/tool-subagent-control/src/list-agents.ts
index 37a4c3bdff..82da2c77f6 100644
--- a/packages/subagent/tool-subagent-control/src/list-agents.ts
+++ b/packages/subagent/tool-subagent-control/src/list-agents.ts
@@ -32,7 +32,7 @@ type ListAgentsEntry =
     readonly kind: 'child'
     readonly id: SessionId
     readonly label: string
-    readonly status: 'running' | 'idle' | 'complete'
+    readonly status: 'running' | 'idle' | 'ready'
     readonly parent?: SessionId
     readonly depth?: number
   }
@@ -52,11 +52,13 @@ function resolveListAgentsRequest(request: ListAgentsRequest): ListAgentsSpec {
 /**
  * Refine one candidate's status through the live Agent registry: `running`
  * for an active driver, `idle` for a resident Agent between turns (possibly
- * waiting on agents it started), and `complete` when no live Agent remains.
+ * waiting on agents it started), and `ready` when no live Agent remains.
+ * `ready` preserves resumability without presenting an inactive conversation
+ * as a terminal result to collect.
  */
-function statusOf(agents: { get(id: SessionId): Agent | undefined }, id: SessionId): 'running' | 'idle' | 'complete' {
+function statusOf(agents: { get(id: SessionId): Agent | undefined }, id: SessionId): 'running' | 'idle' | 'ready' {
   const agent = agents.get(id)
-  if (agent === undefined) return 'complete'
+  if (agent === undefined) return 'ready'
   return agent.status === 'running' ? 'running' : 'idle'
 }
 
@@ -90,10 +92,12 @@ export function apply(ctx: Context): void {
   ctx.tools.register(defineTool({
     name: 'list_agents',
     description:
-      'List your continuable background subagents by durable id and label. Status comes from the live '
+      'List your continuable background subagents by durable id and label. Use it to recall which ones '
+      + 'you started, not to poll for completion — you are told when one finishes. Status comes from the live '
       + 'registry: running means the agent is working right now, idle means it is loaded but between turns '
-      + '(it may be waiting on agents it started), and complete means it exists only in storage — a '
-      + 'direct child remains a `send_message` candidate in every status. The snapshot is not a delivery '
+      + '(it may be waiting on agents it started), and ready means it exists only in storage — resumable, not '
+      + 'terminal, and not a result waiting to be collected; a `send_message` starts a new turn on the same '
+      + 'conversation, and a direct child remains a `send_message` candidate in every status. The snapshot is not a delivery '
       + 'promise — `send_message` performs the authoritative check and may still fail. Children that could '
       + 'not be read are reported as diagnostics instead of being silently dropped. Scope `descendants` '
       + 'walks the whole tree below you in stable pre-order, annotating each entry with its durable direct-parent '
@@ -118,7 +122,7 @@ export function apply(ctx: Context): void {
                 kind: { type: 'string', required: true, enum: ['child'] },
                 id: { type: 'string', required: true },
                 label: { type: 'string', required: true },
-                status: { type: 'string', required: true, enum: ['running', 'idle', 'complete'] },
+                status: { type: 'string', required: true, enum: ['running', 'idle', 'ready'] },
                 parent: { type: 'string' },
                 depth: { type: 'number' },
               },
diff --git a/packages/subagent/tool-subagent-control/tests/list-agents.spec.ts b/packages/subagent/tool-subagent-control/tests/list-agents.spec.ts
index 1c12602dd2..ecdfb0719a 100644
--- a/packages/subagent/tool-subagent-control/tests/list-agents.spec.ts
+++ b/packages/subagent/tool-subagent-control/tests/list-agents.spec.ts
@@ -16,6 +16,7 @@ import type { GenerateOptions, StreamChunk } from '@deepseek-ai/dsh-llm'
 import { LlmAdapter } from '@deepseek-ai/dsh-llm'
 import { MockAdapter, textResponse } from '../../../core/agent-loop/tests/mock-adapter.ts'
 import * as tool from '../src/list-agents.ts'
+import { parkParent } from './park-parent.ts'
 
 /** One scripted response that may wait on a caller-released gate before streaming. */
 interface GatedEntry {
@@ -63,6 +64,7 @@ async function setupWith(adapter: MockAdapter | GatedAdapter) {
   await ctx.plugin(tool)
   ctx.llm.registerAdapter(['mock'], adapter)
   const parent = ctx.agentLoop.create(SessionId('parent'), { provider: 'mock', model: 'mock' })
+  parkParent(ctx, parent)
   return { ctx, parent, adapter }
 }
 
@@ -177,8 +179,10 @@ describe('dsh-tool-subagent-control/list-agents', () => {
     vi.spyOn(ctx.agents, 'get').mockImplementation(id => agents.get(id) as never)
     const result = await callTool(ctx, 'list_agents', {}, parent)
     expect(result.isError).toBe(false)
+    // `ready` is the resumable counterpart to a live `running` record, not a
+    // claim that the child's conversation ended with a result to collect.
     expect(text(result)).toBe(
-      `${started.childId} [complete] — real child\n`
+      `${started.childId} [ready] — real child\n`
       + 'running-child [running] — still working\n'
       + 'waiting-child [idle] — waiting on descendants\n'
       + 'broken-child [diagnostic: corrupt]',
@@ -215,7 +219,21 @@ describe('dsh-tool-subagent-control/list-agents', () => {
     await waitNoActivation(ctx, started.childId)
     const result = await callTool(ctx, 'list_agents', {}, parent)
     expect(result.isError).toBe(false)
-    expect(text(result)).toBe(`${started.childId} [complete] — summarize the doc`)
+    expect(text(result)).toBe(`${started.childId} [ready] — summarize the doc`)
+  })
+
+  it('describes ready as resumable and pins the status vocabulary', async () => {
+    const { ctx } = await setup([])
+    const schema = ctx.tools.schemas().find(candidate => candidate.name === 'list_agents')
+    // Completion reaches the parent through its notice; listing is discovery,
+    // so its inactive status must not send the model looking for a result.
+    expect(schema?.description).toContain('you are told when one finishes')
+    expect(schema?.description).toContain('resumable, not terminal')
+    // The enum is the closed vocabulary the model renders, so pin it rather than
+    // scanning prose that legitimately reads "not to poll for completion".
+    const variants = ctx.tools.get('list_agents')?.output.schema.items?.oneOf ?? []
+    const child = variants.find(variant => variant.properties?.kind?.enum?.includes('child'))
+    expect(child?.properties?.status?.enum).toEqual(['running', 'idle', 'ready'])
   })
 
   it('fails loud when invoked without a calling agent', async () => {
@@ -321,7 +339,7 @@ describe('dsh-tool-subagent-control/list-agents', () => {
     const result = await callTool(ctx, 'list_agents', { scope: 'descendants' }, parent)
     expect(result.isError).toBe(false)
     expect(text(result)).toBe(
-      'deep-leaf [complete] parent=one-shot-mid depth=2 — deep leaf\n'
+      'deep-leaf [ready] parent=one-shot-mid depth=2 — deep leaf\n'
       + `broken-node [diagnostic: unavailable] parent=${parent.id} depth=1`,
     )
   })
diff --git a/packages/subagent/tool-subagent-control/tests/park-parent.ts b/packages/subagent/tool-subagent-control/tests/park-parent.ts
new file mode 100644
index 0000000000..7f61f55d51
--- /dev/null
+++ b/packages/subagent/tool-subagent-control/tests/park-parent.ts
@@ -0,0 +1,22 @@
+/**
+ * Shared suite helper: keep this package's stand-in parent out of a scripted
+ * model corpus.
+ * @module park-parent
+ */
+
+import type { Context } from '@deepseek-ai/cordis'
+import type { SessionId } from '@deepseek-ai/dsh-session'
+
+/**
+ * Reject every step of the stand-in parent. Each child settlement wakes its
+ * parent, and these suites size their scripts for child turns only; the tests
+ * assert on delivery rather than on the parent's own turn.
+ * @param ctx - the booted test context.
+ * @param parent - the stand-in parent whose turns must not reach the model.
+ */
+export function parkParent(ctx: Context, parent: { id: SessionId }): void {
+  ctx.on('agent/pre-step', async ({ agent: subject }, next) => {
+    if (subject.id !== parent.id) return next()
+    return { kind: 'reject' as const }
+  })
+}
diff --git a/packages/subagent/tool-subagent-control/tests/tool-subagent-control.spec.ts b/packages/subagent/tool-subagent-control/tests/tool-subagent-control.spec.ts
index 4c378fcbc6..4960992068 100644
--- a/packages/subagent/tool-subagent-control/tests/tool-subagent-control.spec.ts
+++ b/packages/subagent/tool-subagent-control/tests/tool-subagent-control.spec.ts
@@ -15,6 +15,7 @@ import type { GenerateOptions, StreamChunk } from '@deepseek-ai/dsh-llm'
 import { LlmAdapter } from '@deepseek-ai/dsh-llm'
 import { MockAdapter, textResponse } from '../../../core/agent-loop/tests/mock-adapter.ts'
 import * as tool from '../src/index.ts'
+import { parkParent } from './park-parent.ts'
 
 /** One scripted response that may wait on a caller-released gate before streaming. */
 interface GatedEntry {
@@ -62,6 +63,7 @@ async function setupWith(adapter: MockAdapter | GatedAdapter) {
   await ctx.plugin(tool)
   ctx.llm.registerAdapter(['mock'], adapter)
   const parent = ctx.agentLoop.create(SessionId('parent'), { provider: 'mock', model: 'mock' })
+  parkParent(ctx, parent)
   return { ctx, parent, adapter }
 }
 
diff --git a/packages/subagent/tool-subagent-report/README.i18n.yaml b/packages/subagent/tool-subagent-report/README.i18n.yaml
index 8d64d7e87f..13a892fa8a 100644
--- a/packages/subagent/tool-subagent-report/README.i18n.yaml
+++ b/packages/subagent/tool-subagent-report/README.i18n.yaml
@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write packages/subagent/tool-subagent-report/README.md
-README.md: 1772b72d83563e89ed0e3814e515b06a2b36fe0f
-README.zh.md: 94101fbc9611a04c56ed3a9171889499c9144c21
+README.md: b38b6b541d9c03174de1515df7a0babe5da055b4
+README.zh.md: 0c5c063eded1e9057ee8886f03dd6494dd9eeed8
diff --git a/packages/subagent/tool-subagent-report/README.md b/packages/subagent/tool-subagent-report/README.md
index 1772b72d83..b38b6b541d 100644
--- a/packages/subagent/tool-subagent-report/README.md
+++ b/packages/subagent/tool-subagent-report/README.md
@@ -2,15 +2,15 @@
 
 English | [中文](README.zh.md)
 
-The optional child-scoped `report` tool is a thin adapter over `ctx.subagents.reportFrom()`. It gives every continuable in-process child a return channel to the Agent that started it. The package registers a continuable-child setup contribution instead of a global tool, so `report` exists only inside those children. Roots, one-shot subagents, remote subagent providers, sibling scopes, and agentless tool execution never present or execute it. Installing this package grants only that child-scoped capability; the parent-to-child direction remains the independent [`@deepseek-ai/dsh-tool-subagent-control`](../tool-subagent-control/README.md), and continuable mode depends on neither package.
+The optional child-scoped `report` tool is a thin adapter over `ctx.subagents.reportFrom()`. It gives every continuable in-process child a return channel to the Agent that started it, and installs the prompt section that instructs the child to use it. The package registers a continuable-child setup contribution instead of a global tool, so the tool and its guidance exist only inside those children. Roots, one-shot subagents, remote subagent providers, sibling scopes, and agentless tool execution never present or execute it. Installing this package grants only that child-scoped capability; the parent-to-child direction remains the independent [`@deepseek-ai/dsh-tool-subagent-control`](../tool-subagent-control/README.md), and continuable mode depends on neither package.
 
-A child may call `report` zero or many times in one turn. A successful call neither concludes the turn, settles the Activation, nor prevents later parent follow-ups, and finishing a turn never reports automatically. The tool accepts no recipient: `exec.agent` is the sender's exact live Agent and the authority credential, and the service derives the sole recipient from that child's durable `parentSession`. Success returns the stable `MessageId` of the parent-accepted message, not a read receipt, an inbox-occurrence id, a parent-log acknowledgement, a turn-completion receipt, or a persistence flush. A parent absent from the registry fails the call with `direct parent is not live; report was not delivered` — registry presence governs parent resolution, and a registered parent already in host-owned disposal still accepts while its log admits appends. The service performs no injection, parent cold resume, or offline mailbox write; the durable child transcript remains the recovery source, and a failed tool call does not prove non-delivery (a later `tools/post-execute` veto can fail a call whose report was already accepted).
+The child-scoped `tool:report` prompt section instructs the child to call `report` once before finishing, with a self-contained answer, and earlier whenever a partial finding changes what the parent should do next. The instruction is guidance, not enforcement: the mechanism still accepts zero or many calls in one turn, and no runtime path rejects a child that never reports. A successful call neither concludes the turn, settles the Activation, nor prevents later parent follow-ups, and finishing a turn never reports automatically. The tool accepts no recipient: `exec.agent` is the sender's exact live Agent and the authority credential, and the service derives the sole recipient from that child's durable `parentSession`. Success returns the stable `MessageId` of the parent-accepted message, not a read receipt, an inbox-occurrence id, a parent-log acknowledgement, a turn-completion receipt, or a persistence flush. A parent absent from the registry fails the call with `direct parent is not live; report was not delivered` — registry presence governs parent resolution, and a registered parent already in host-owned disposal still accepts while its log admits appends. The service performs no injection, parent cold resume, or offline mailbox write; the durable child transcript remains the recovery source, and a failed tool call does not prove non-delivery (a later `tools/post-execute` veto can fail a call whose report was already accepted).
 
-`reportDelivery` selects parent scheduling for every accepted report. `quiet` (the default) uses `parent.inject()`, adding model-facing context without starting a parent model request: an idle parent's append completes before the call returns, while a report reaching an admitting or running parent stages for the next safe log position. `wakeup` uses `parent.followup()`, creating exactly one ordinary later parent turn and waking a parked parent driver; it never steers an open turn. This is deployment scheduling policy, so the model-facing schema cannot select or override it per call.
+`reportDelivery` selects parent scheduling for every accepted report. `wakeup` (the default) uses `parent.followup()`, creating exactly one ordinary later parent turn and waking a parked parent driver; it never steers an open turn. It is the default because a parent that already parked has no other reason to look, so quiet delivery would leave an accepted report unread until something unrelated woke it. `quiet` uses `parent.inject()`, adding model-facing context without starting a parent model request: an idle parent's append completes before the call returns, while a report reaching an admitting or running parent stages for the next safe log position. This is deployment scheduling policy, so the model-facing schema cannot select or override it per call.
 
 Scope-local registration deliberately survives the child's global `toolFilter`, so a delegation allow-list cannot remove the only return channel. A deployment that requires a child with no return channel omits this package.
 
-The contribution body is exported as `installReportTool(childCtx, ctx, delivery)` so inspection consumers can install `report` into a minted child scope. The generated tool catalog uses that path because the global registry cannot expose a scope-local schema. Production composition still enters through `apply()`; the subagent seam's contribution registry remains private.
+The contribution body is exported as `installReportTool(childCtx, ctx, delivery)` so inspection consumers can install `report` and its guidance into a minted child scope, and returns the one disposer revoking both. The generated tool catalog uses that path because the global registry cannot expose a scope-local schema. Production composition still enters through `apply()`; the subagent seam's contribution registry remains private.
 
 ## Model Experience
 
@@ -18,15 +18,15 @@ The contribution body is exported as `installReportTool(childCtx, ctx, delivery)
 
 #### What the model sees
 
-The generated [`report` schema](../../../docs/tool-catalog.md#deepseek-aidsh-tool-subagent-report): one required `output` string. Its description states that reporting is explicit and repeatable, reaches only the Agent that started the child, and does not end the turn. It carries no recipient or delivery-mode parameter.
+The generated [`report` schema](../../../docs/tool-catalog.md#deepseek-aidsh-tool-subagent-report): one required `output` string. Its description states that the child must report once before finishing, that reporting reaches only the Agent that started the child, and that it does not end the turn. It carries no recipient or delivery-mode parameter. The separate `tool:report` prompt section repeats the obligation outside the schema, where a child that ignores tool descriptions still reads it.
 
 #### Token effect
 
-Fixed schema cost per continuable-child request, and none in any other Agent's requests.
+Fixed schema and prompt-section cost per continuable-child request, and none in any other Agent's requests.
 
 #### KV Cache effect
 
-Prefix-stable within a child; the schema does not change at runtime. Removing the package revokes the schema from resident children, which changes their next request prefix.
+Prefix-stable within a child; neither the schema nor the section changes at runtime. Removing the package revokes both from resident children, which changes their next request prefix.
 
 ### Report result
 
@@ -36,7 +36,7 @@ Prefix-stable within a child; the schema does not change at runtime. Removing th
 
 #### Token effect
 
-One short acknowledgement per call in the reporting child. The reported content is additionally billed to the parent: quiet delivery adds it to the parent's next request, while waking delivery makes it the sole ordinary message of one new parent turn.
+One short acknowledgement per call in the reporting child. The reported content is additionally billed to the parent: waking delivery makes it the sole ordinary message of one new parent turn, while quiet delivery adds it to the parent's next request.
 
 #### KV Cache effect
 
@@ -61,6 +61,6 @@ Append-only; the report follows the parent's reusable request prefix. Waking del
 - **A parent whose host-owned disposal already started can still accept** — `AgentHandle.dispose()` cancels, awaits quiescence, and only then unwinds the scope and leaves the registry; it exposes no signal for "disposal started." A report accepted in that window is appended to the parent's transcript, but that parent will not act on it in this process. A continuation-manager-owned parent rejects forest teardown through the manager's admission boundary.
 - **Acceptance is weaker than durable delivery** — there is no durable mailbox, idempotency key, delivery receipt, retry protocol, or exactly-once claim. A process failure after one side recorded acceptance leaves the outcome ambiguous, and an external retry may duplicate the report.
 - **A staged quiet report is not immediately reconstructable** — acceptance returns its stable `MessageId`, but the parent Session reconstructs the framed content only after pending context reaches its ordinary log boundary.
-- **Granting waits for the next Activation; revocation is immediate** — installing this package after a child becomes resident grants `report` only on that child's next Activation, while removing the package revokes the schema from resident children immediately.
+- **Granting waits for the next Activation; revocation is immediate** — installing this package after a child becomes resident grants `report` and its guidance only on that child's next Activation, while removing the package revokes both from resident children immediately.
 - **Nested reporting reaches exactly one edge upward** — a grandchild reports to its direct child parent, never to the top-level coordinator, which must explicitly report a derived update later.
-- **No rate limiting** — `wakeup` mode can amplify model work when nested children report frequently; the deployment owns that choice by selecting the mode.
+- **No rate limiting** — the default `wakeup` mode can amplify model work when nested children report frequently; a deployment that accepts unread reports over that amplification selects `quiet`.
diff --git a/packages/subagent/tool-subagent-report/README.zh.md b/packages/subagent/tool-subagent-report/README.zh.md
index 94101fbc96..0c5c063ede 100644
--- a/packages/subagent/tool-subagent-report/README.zh.md
+++ b/packages/subagent/tool-subagent-report/README.zh.md
@@ -2,15 +2,15 @@
 
 [English](README.md) | 中文
 
-可选的子级作用域 `report` 工具是 `ctx.subagents.reportFrom()` 之上的轻量适配器。它为每个可继续的进程内子级提供一条返回通道，指向启动该子级的 Agent（智能体）。本包注册的是可继续子级设置贡献，而不是全局工具，因此 `report` 只存在于这些子级内部。根 Agent、一次性 subagent、远程 subagent 提供方、同级作用域以及不关联 Agent 的工具执行都不会提供或执行它。安装本包只授予这项子级作用域能力；父到子方向仍由独立的 [`@deepseek-ai/dsh-tool-subagent-control`](../tool-subagent-control/README.md) 负责，可继续模式不依赖这两个包中的任一个。
+可选的子级作用域 `report` 工具是 `ctx.subagents.reportFrom()` 之上的轻量适配器。它为每个可继续的进程内子级提供一条返回通道，指向启动该子级的 Agent（智能体），并安装指示子级使用该通道的提示词 section。本包注册的是可继续子级设置贡献，而不是全局工具，因此该工具及其指引只存在于这些子级内部。根 Agent、一次性 subagent、远程 subagent 提供方、同级作用域以及不关联 Agent 的工具执行都不会提供或执行它。安装本包只授予这项子级作用域能力；父到子方向仍由独立的 [`@deepseek-ai/dsh-tool-subagent-control`](../tool-subagent-control/README.md) 负责，可继续模式不依赖这两个包中的任一个。
 
-子级可以在一个轮次中调用 `report` 零次或多次。调用成功既不会结束轮次或结算 Activation，也不会阻止父级后续消息；轮次结束也绝不会自动上报。该工具不接受接收方参数：`exec.agent` 是发送方确切在线的 Agent，也是权限凭据；服务根据该子级持久化的 `parentSession` 推导唯一接收方。成功时返回父级已接受消息的稳定 `MessageId`，不表示已读回执、inbox 中该次出现的 id、父级日志确认、轮次完成回执或持久化刷盘。父级解析由注册表中的存在性决定：父级不在注册表时，调用失败并返回 `direct parent is not live; report was not delivered`；已开始由宿主管理的 dispose（资源释放）但仍在注册表中的父级在其日志仍接受追加时仍会接受。服务不会执行注入、父级冷恢复或离线 mailbox 写入；持久化子级 transcript（文本记录）仍是恢复依据，且工具调用失败不能证明未送达（后续 `tools/post-execute` 否决可能让报告已被接受的调用以失败结束）。
+子级作用域的 `tool:report` 提示词 section 要求子级在结束前调用一次 `report` 并给出自足的答案，并在部分发现会改变父级下一步动作时提前上报。该指令是引导而非强制：机制本身仍接受一个轮次中调用零次或多次，也没有任何运行时路径会拒绝从不上报的子级。调用成功既不会结束轮次或结算 Activation，也不会阻止父级后续消息；轮次结束也绝不会自动上报。该工具不接受接收方参数：`exec.agent` 是发送方确切在线的 Agent，也是权限凭据；服务根据该子级持久化的 `parentSession` 推导唯一接收方。成功时返回父级已接受消息的稳定 `MessageId`，不表示已读回执、inbox 中该次出现的 id、父级日志确认、轮次完成回执或持久化刷盘。父级解析由注册表中的存在性决定：父级不在注册表时，调用失败并返回 `direct parent is not live; report was not delivered`；已开始由宿主管理的 dispose（资源释放）但仍在注册表中的父级在其日志仍接受追加时仍会接受。服务不会执行注入、父级冷恢复或离线 mailbox 写入；持久化子级 transcript（文本记录）仍是恢复依据，且工具调用失败不能证明未送达（后续 `tools/post-execute` 否决可能让报告已被接受的调用以失败结束）。
 
-`reportDelivery` 为每条已接受的报告选择父级调度方式。`quiet`（默认值）使用 `parent.inject()`，在不启动父级模型请求的情况下添加面向模型的上下文：父级空闲时，追加操作会在调用返回前完成；报告到达正在准入或运行的父级时，则会暂存到下一个安全日志位置。`wakeup` 使用 `parent.followup()`，恰好创建一个普通的后续父级轮次，并唤醒停驻的父级驱动；它绝不会对正在运行的轮次进行 steering（中途引导）。这是部署调度策略，因此面向模型的 schema 不能在单次调用中选择或覆盖该策略。
+`reportDelivery` 为每条已接受的报告选择父级调度方式。`wakeup`（默认值）使用 `parent.followup()`，恰好创建一个普通的后续父级轮次，并唤醒停驻的父级驱动；它绝不会对正在运行的轮次进行 steering（中途引导）。之所以作为默认值：已经停驻的父级没有别的理由再去查看，静默投递会让一条已被接受的报告一直无人阅读，直到别的事件把父级唤醒。`quiet` 使用 `parent.inject()`，在不启动父级模型请求的情况下添加面向模型的上下文：父级空闲时，追加操作会在调用返回前完成；报告到达正在准入或运行的父级时，则会暂存到下一个安全日志位置。这是部署调度策略，因此面向模型的 schema 不能在单次调用中选择或覆盖该策略。
 
 作用域局部注册有意不受子级全局 `toolFilter` 影响，因此委派允许列表无法移除唯一的返回通道。需要子级不具备返回通道的部署应省略本包。
 
-贡献体以 `installReportTool(childCtx, ctx, delivery)` 导出，以便检查类消费方把 `report` 安装到新创建的子级作用域中。全局注册表无法公开作用域局部 schema，因此生成的工具目录会使用这条路径。生产组合仍通过 `apply()` 进入；subagent seam 的贡献注册表保持私有。
+贡献体以 `installReportTool(childCtx, ctx, delivery)` 导出，以便检查类消费方把 `report` 及其指引安装到新创建的子级作用域中，并返回同时撤销两者的唯一 disposer。全局注册表无法公开作用域局部 schema，因此生成的工具目录会使用这条路径。生产组合仍通过 `apply()` 进入；subagent seam 的贡献注册表保持私有。
 
 ## 模型体验
 
@@ -18,15 +18,15 @@
 
 #### 模型看到的内容
 
-已生成的 [`report` schema](../../../docs/tool-catalog.md#deepseek-aidsh-tool-subagent-report)：包含一个必填 `output` 字符串。其描述说明上报需要显式调用且可以重复，只会到达启动该子级的 Agent，并且不会结束轮次。它不包含接收方或投递模式参数。
+已生成的 [`report` schema](../../../docs/tool-catalog.md#deepseek-aidsh-tool-subagent-report)：包含一个必填 `output` 字符串。其描述说明子级必须在结束前上报一次，上报只会到达启动该子级的 Agent，并且不会结束轮次。它不包含接收方或投递模式参数。独立的 `tool:report` 提示词 section 在 schema 之外重申该义务，使忽略工具描述的子级仍能读到。
 
 #### Token 影响
 
-每个可继续子级请求支付固定的 schema 成本，其他任何 Agent 的请求均无此成本。
+每个可继续子级请求支付固定的 schema 与提示词 section 成本，其他任何 Agent 的请求均无此成本。
 
 #### KV Cache 影响
 
-子级中的前缀保持稳定；schema 不会在运行时改变。移除本包会从驻留子级中撤销该 schema，从而改变其下一次请求前缀。
+子级中的前缀保持稳定；schema 与该 section 都不会在运行时改变。移除本包会从驻留子级中撤销两者，从而改变其下一次请求前缀。
 
 ### 上报结果
 
@@ -36,7 +36,7 @@
 
 #### Token 影响
 
-每次调用都会在执行上报的子级中产生一条简短确认消息。父级还会为上报内容支付 token 成本：静默投递会把内容加入父级的下一次请求，唤醒投递则会使该内容成为一个新父级轮次中唯一的普通消息。
+每次调用都会在执行上报的子级中产生一条简短确认消息。父级还会为上报内容支付 token 成本：唤醒投递会使该内容成为一个新父级轮次中唯一的普通消息，静默投递则把内容加入父级的下一次请求。
 
 #### KV Cache 影响
 
@@ -61,6 +61,6 @@
 - **父级可能在宿主启动 dispose 后继续接受报告**：`AgentHandle.dispose()` 会先取消并等待完全停稳，然后才撤销作用域并离开注册表；它不公开「dispose 已开始」信号。在该窗口内接受的报告会追加到父级 transcript，但该父级不会在本进程中处理它。对于由继续执行管理器拥有的父级，管理器的准入边界会在整片森林拆卸期间拒绝该上报。
 - **接受弱于持久投递**：没有持久化 mailbox、幂等键、投递回执、重试协议，也不保证恰好一次。任一侧记录接受后若进程失败，结果都不明确；外部重试可能产生重复上报。
 - **暂存的静默报告无法立即重建**：接受时会返回其稳定 `MessageId`，但只有当待处理上下文到达普通日志边界后，父级 Session 才能重建带前缀的内容。
-- **授权须等到下一个 Activation，撤销则立即生效**：子级驻留后再安装本包，只会在该子级的下一个 Activation 中授予 `report`；移除本包则会立即从驻留子级撤销该 schema。
+- **授权须等到下一个 Activation，撤销则立即生效**：子级驻留后再安装本包，只会在该子级的下一个 Activation 中授予 `report` 及其指引；移除本包则会立即从驻留子级撤销两者。
 - **嵌套上报只向上到达一条直接边**：孙级只向作为其直接父级的子级上报，不会直接到达顶层协调器；该直接父级必须随后显式发出一条衍生更新。
-- **没有速率限制**：嵌套子级频繁上报时，`wakeup` 模式会放大模型工作量；部署通过选择模式自行承担这一取舍。
+- **没有速率限制**：嵌套子级频繁上报时，默认的 `wakeup` 模式会放大模型工作量；宁可接受报告无人阅读也要避免这种放大的部署应选择 `quiet`。
diff --git a/packages/subagent/tool-subagent-report/package.json b/packages/subagent/tool-subagent-report/package.json
index f9dcd90588..8c2eb5d999 100644
--- a/packages/subagent/tool-subagent-report/package.json
+++ b/packages/subagent/tool-subagent-report/package.json
@@ -35,6 +35,7 @@
     "@deepseek-ai/dsh-invariants": "workspace:^",
     "@deepseek-ai/dsh-llm": "workspace:^",
     "@deepseek-ai/dsh-subagent": "workspace:^",
+    "@deepseek-ai/dsh-system-prompt": "workspace:^",
     "@deepseek-ai/dsh-tools": "workspace:^",
     "@deepseek-ai/cordis": "workspace:^"
   },
@@ -52,6 +53,7 @@
     "@deepseek-ai/dsh-session-persistence-jsonl": "workspace:^",
     "@deepseek-ai/dsh-subagent": "workspace:^",
     "@deepseek-ai/dsh-subagent-spawn": "workspace:^",
+    "@deepseek-ai/dsh-system-prompt": "workspace:^",
     "@deepseek-ai/dsh-tool-subagent-control": "workspace:^",
     "@deepseek-ai/dsh-tools": "workspace:^",
     "@deepseek-ai/cordis": "workspace:^"
diff --git a/packages/subagent/tool-subagent-report/src/index.ts b/packages/subagent/tool-subagent-report/src/index.ts
index b33a5a29ab..6f2a6b3c6a 100644
--- a/packages/subagent/tool-subagent-report/src/index.ts
+++ b/packages/subagent/tool-subagent-report/src/index.ts
@@ -1,7 +1,7 @@
 /**
- * The child-scoped `report` tool, installed into every continuable in-process
- * child's unpublished context. Roots, one-shot children, remote providers, and
- * agentless executions never see the registration.
+ * The child-scoped `report` tool and its usage guidance, installed into every
+ * continuable in-process child's unpublished context. Roots, one-shot children,
+ * remote providers, and agentless executions never see the registration.
  *
  * @module @deepseek-ai/dsh-tool-subagent-report
  */
@@ -11,86 +11,131 @@ import z from '@deepseek-ai/schemastery'
 import type { Agent } from '@deepseek-ai/dsh-agent'
 import type { ContentBlock } from '@deepseek-ai/dsh-llm'
 import type { SubagentReportDelivery } from '@deepseek-ai/dsh-subagent'
+import type {} from '@deepseek-ai/dsh-system-prompt'
 import { defineTool } from '@deepseek-ai/dsh-tools'
 
 export const name = 'tool-subagent-report'
-// The contribution registers only through childCtx.tools, but declaring tools
-// makes Loader ordering fail at load instead of the next child materialization.
-export const inject = ['subagents', 'tools']
+// The contribution registers only through childCtx.tools and
+// childCtx.systemPrompt, but declaring both services makes Loader ordering fail
+// at load instead of at the next child materialization.
+export const inject = ['subagents', 'tools', 'systemPrompt']
+
+/** Guidance order after every per-tool section a continuable child can carry. */
+const REPORT_SECTION_ORDER = 117
 
 /** Config: how accepted reports are scheduled on the parent. */
 export interface Config {
   /**
-   * Parent scheduling (default `quiet`). `quiet` adds context without waking;
-   * `wakeup` creates one ordinary later parent turn.
+   * Parent scheduling (default `wakeup`). `wakeup` creates one ordinary later
+   * parent turn; `quiet` adds context without waking, so a parked parent learns
+   * of the report only when something else wakes it.
    */
   reportDelivery?: SubagentReportDelivery
 }
 
 export const Config: z<Config> = z.object({
-  reportDelivery: z.union(['quiet', 'wakeup'] as const).default('quiet'),
+  reportDelivery: z.union(['quiet', 'wakeup'] as const).default('wakeup'),
 })
 
 /**
- * Install `report` into one continuable child's scope.
- * @param childCtx - child-scoped context receiving the tool.
+ * Install `report` and its usage guidance into one continuable child's scope.
+ * Both registrations are owned by that scope and are therefore invisible to the
+ * child's parent and siblings.
+ * @param childCtx - child-scoped context receiving the tool and the guidance.
  * @param ctx - service context used for delivery.
  * @param delivery - resolved deployment scheduling policy.
- * @returns disposer for this one registration.
+ * @returns disposer that attempts both child registrations before reporting cleanup failures.
  */
 export function installReportTool(
   childCtx: Context,
   ctx: Context,
   delivery: SubagentReportDelivery,
 ): () => void {
-  return childCtx.tools.register(defineTool({
-    name: 'report',
-    description:
-      'Report selected content to the agent that started you. Call this zero or more times for progress, '
-      + 'findings, or a final answer. Reporting does not end your turn or finish your work, and only your '
-      + 'direct parent receives it. A failed call may still have arrived, so do not blindly repeat it.',
-    parameters: {
-      output: {
-        type: 'string',
-        required: true,
-        description: 'Self-contained content for your parent; it does not see your private work.',
-      },
-    },
-    output: {
-      schema: {
-        type: 'object',
-        additionalProperties: false,
-        properties: {
-          messageId: { type: 'string', required: true },
+  const disposeSection = childCtx.systemPrompt.section({
+    name: 'tool:report',
+    order: REPORT_SECTION_ORDER,
+    text: 'Deliver your result with the report tool before you finish: call it once with a self-contained '
+      + 'answer. The agent that started you shares your workspace but does not automatically receive your '
+      + 'transcript, tool output, or reasoning, so a closing remark such as "done" leaves it nothing it can '
+      + 'use. Report earlier as well whenever a partial finding changes what that agent should do next; '
+      + 'reporting never ends your turn.',
+  })
+  let disposeTool: () => void
+  try {
+    disposeTool = childCtx.tools.register(defineTool({
+      name: 'report',
+      description:
+        'Report selected content to the agent that started you. Call this once before you finish, with a '
+        + 'self-contained final result, and earlier for progress or findings that change what that agent does '
+        + 'next. That agent shares your workspace but does not automatically receive your transcript, tool '
+        + 'output, or reasoning, so finishing your work is not itself a result. Reporting does not end your '
+        + 'turn or finish your work, and only your direct parent receives it. A failed call may still have '
+        + 'arrived, so do not blindly repeat it.',
+      parameters: {
+        output: {
+          type: 'string',
+          required: true,
+          description: 'Actionable content for your parent; summarize conclusions and reference relevant shared paths.',
         },
       },
-      render: (_args, value) => [{
-        type: 'text',
-        text: `report accepted by the agent that started you as message ${value.messageId}`,
-      }],
-    },
-    async execute(args, exec) {
-      const content: ContentBlock[] = [{ type: 'text', text: args.output }]
-      // Scope-local resolution guarantees an Agent. The service still verifies
-      // its exact live Activation identity at the authority boundary.
-      const messageId = await ctx.subagents.reportFrom(exec.agent as Agent, content, {
-        delivery,
-        signal: exec.signal,
-      })
-      return { messageId }
-    },
-  }))
+      output: {
+        schema: {
+          type: 'object',
+          additionalProperties: false,
+          properties: {
+            messageId: { type: 'string', required: true },
+          },
+        },
+        render: (_args, value) => [{
+          type: 'text',
+          text: `report accepted by the agent that started you as message ${value.messageId}`,
+        }],
+      },
+      async execute(args, exec) {
+        const content: ContentBlock[] = [{ type: 'text', text: args.output }]
+        // Scope-local resolution guarantees an Agent. The service still verifies
+        // its exact live Activation identity at the authority boundary.
+        const messageId = await ctx.subagents.reportFrom(exec.agent as Agent, content, {
+          delivery,
+          signal: exec.signal,
+        })
+        return { messageId }
+      },
+    }))
+  } catch (error: unknown) {
+    try {
+      disposeSection()
+    } catch (rollbackError: unknown) {
+      throw new AggregateError(
+        [error, rollbackError],
+        'failed to register the report tool and roll back its prompt guidance',
+      )
+    }
+    throw error
+  }
+  return () => {
+    const failures: unknown[] = []
+    for (const dispose of [disposeTool, disposeSection]) {
+      try {
+        dispose()
+      } catch (error: unknown) {
+        failures.push(error)
+      }
+    }
+    if (failures.length > 0) {
+      throw new AggregateError(failures, 'failed to revoke report tool and prompt registrations')
+    }
+  }
 }
 
 /**
  * Register the continuable-child contribution.
- * @param ctx - context carrying tools and the subagent service.
+ * @param ctx - context carrying tools, the system prompt, and the subagent service.
  * @param config - deployment scheduling policy.
  */
 export function apply(ctx: Context, config: Config = {}): void {
-  // Config() applies the schema default ('quiet') at runtime; the schemastery
-  // return type keeps the input's optional shape, so assert the resolved
-  // shape here — no runtime fallback exists or is wanted.
+  // Config() applies the schema default at runtime; the schemastery return
+  // type keeps the input's optional shape, so assert the resolved one.
   const { reportDelivery } = Config(config) as { reportDelivery: SubagentReportDelivery }
   ctx.subagents.registerContinuableSetup(childCtx =>
     installReportTool(childCtx, ctx, reportDelivery))
diff --git a/packages/subagent/tool-subagent-report/tests/tool-subagent-report.spec.ts b/packages/subagent/tool-subagent-report/tests/tool-subagent-report.spec.ts
index f33901c8c7..ac106a5680 100644
--- a/packages/subagent/tool-subagent-report/tests/tool-subagent-report.spec.ts
+++ b/packages/subagent/tool-subagent-report/tests/tool-subagent-report.spec.ts
@@ -4,6 +4,7 @@ import { tmpdir } from 'node:os'
 import { join } from 'node:path'
 import { Context } from '@deepseek-ai/cordis'
 import type { Agent } from '@deepseek-ai/dsh-agent'
+import { assembleContextFor } from '@deepseek-ai/dsh-agent'
 import AgentLoop from '@deepseek-ai/dsh-agent-loop'
 import { mountAgentLoopTestDependencies } from '@deepseek-ai/dsh-agent-loop-testkit'
 import { CallId, LlmAdapter } from '@deepseek-ai/dsh-llm'
@@ -96,6 +97,17 @@ function callReport(ctx: Context, child: Agent, output: string, signal = testSig
   })
 }
 
+/** Occupy the child-local report name to force installation rollback. */
+function registerReportConflict(child: Agent): () => void {
+  return child.ctx.tools.register({
+    name: 'report',
+    description: 'conflicting report fixture',
+    parameters: { type: 'object', properties: {} },
+    output: { schema: { type: 'object', properties: {} }, render: () => [] },
+    execute: () => Promise.resolve({}),
+  })
+}
+
 /** Reports already visible or still pending in one Agent. */
 function reports(agent: Agent): { id: string; text: string; sender: string }[] {
   const visible = agent.session.events.flatMap(event => event.type === 'user/message' ? [event.data] : [])
@@ -113,6 +125,12 @@ function renderedText(result: { content: { type: string; text?: string }[] }): s
   return result.content.flatMap(block => block.type === 'text' ? [block.text ?? ''] : []).join('')
 }
 
+/** The prompt sections one agent's scope assembles, by name. */
+async function sectionNames(ctx: Context, agent: Agent): Promise<string[]> {
+  const assembly = await ctx.systemPrompt.assemble(assembleContextFor(agent))
+  return assembly.sections.map(section => section.name)
+}
+
 describe('dsh-tool-subagent-report', () => {
   it('registers report only in continuable child scopes', async () => {
     const { ctx, parent } = await setup()
@@ -309,16 +327,101 @@ describe('dsh-tool-subagent-report', () => {
     const { ctx, parent, fiber } = await setup()
     const { child } = await startChild(ctx, parent)
     expect(ctx.tools.schemas(child).map(schema => schema.name)).toContain('report')
+    expect(await sectionNames(ctx, child)).toContain('tool:report')
 
     await fiber?.dispose()
     expect(ctx.tools.schemas(child).map(schema => schema.name)).not.toContain('report')
+    expect(await sectionNames(ctx, child)).not.toContain('tool:report')
     expect((await callReport(ctx, child, 'revoked')).isError).toBe(true)
 
     const late = await ctx.plugin(tool, { reportDelivery: 'quiet' })
     expect(ctx.tools.schemas(child).map(schema => schema.name)).not.toContain('report')
+    expect(await sectionNames(ctx, child)).not.toContain('tool:report')
     await late.dispose()
   })
 
+  it('rolls back prompt guidance when tool registration fails', async () => {
+    const { ctx, parent } = await setup({ load: false })
+    const { child } = await startChild(ctx, parent)
+    const disposeConflict = registerReportConflict(child)
+
+    expect(() => tool.installReportTool(child.ctx, ctx, 'quiet')).toThrow(/already registered in this scope/)
+    expect(await sectionNames(ctx, child)).not.toContain('tool:report')
+    disposeConflict()
+  })
+
+  it('aggregates a registration failure with a prompt rollback failure', async () => {
+    const { ctx, parent } = await setup({ load: false })
+    const { child } = await startChild(ctx, parent)
+    const disposeConflict = registerReportConflict(child)
+    const rollbackFailure = new Error('prompt rollback listener failed')
+    let promptChanges = 0
+    const off = ctx.on('system-prompt/change', () => {
+      promptChanges++
+      if (promptChanges === 2) throw rollbackFailure
+    })
+
+    let failure: unknown
+    try {
+      tool.installReportTool(child.ctx, ctx, 'quiet')
+    } catch (error: unknown) {
+      failure = error
+    }
+    off()
+
+    expect(failure).toBeInstanceOf(AggregateError)
+    if (!(failure instanceof AggregateError)) throw new Error('expected aggregate installation failure')
+    expect(failure.errors).toHaveLength(2)
+    expect(String(failure.errors[0])).toContain('already registered in this scope')
+    expect(failure.errors[1]).toBe(rollbackFailure)
+    expect(await sectionNames(ctx, child)).not.toContain('tool:report')
+    disposeConflict()
+  })
+
+  it('attempts both revocations and aggregates change-listener failures', async () => {
+    const { ctx, parent } = await setup({ load: false })
+    const { child } = await startChild(ctx, parent)
+    const dispose = tool.installReportTool(child.ctx, ctx, 'quiet')
+    const toolFailure = new Error('tool removal listener failed')
+    const promptFailure = new Error('prompt removal listener failed')
+    const offTool = ctx.on('tools/change', () => { throw toolFailure })
+    const offPrompt = ctx.on('system-prompt/change', () => { throw promptFailure })
+
+    let failure: unknown
+    try {
+      dispose()
+    } catch (error: unknown) {
+      failure = error
+    }
+    offPrompt()
+    offTool()
+
+    expect(failure).toBeInstanceOf(AggregateError)
+    if (!(failure instanceof AggregateError)) throw new Error('expected aggregate revocation failure')
+    expect(failure.errors).toEqual([toolFailure, promptFailure])
+    expect(ctx.tools.schemas(child).map(schema => schema.name)).not.toContain('report')
+    expect(await sectionNames(ctx, child)).not.toContain('tool:report')
+  })
+
+  it('scopes the report guidance to the child that owns it', async () => {
+    const { ctx, parent } = await setup()
+    const { child } = await startChild(ctx, parent, 'first child')
+    const { child: sibling } = await startChild(ctx, parent, 'second child')
+
+    const assembly = await ctx.systemPrompt.assemble(assembleContextFor(child))
+    const guidance = assembly.sections.find(section => section.name === 'tool:report')
+    // Pins the model-visible instruction that makes the return channel a
+    // contract rather than an option the child may quietly skip.
+    expect(guidance?.text).toContain('Deliver your result with the report tool before you finish')
+    expect(guidance?.text).toContain('reporting never ends your turn')
+
+    expect(await sectionNames(ctx, parent)).not.toContain('tool:report')
+    // A sibling installs its own copy; neither child can observe the other's.
+    expect(await sectionNames(ctx, sibling)).toContain('tool:report')
+    expect((await ctx.systemPrompt.assemble()).sections.map(section => section.name))
+      .not.toContain('tool:report')
+  })
+
   it('rolls back materialization when a setup contribution revokes itself', async () => {
     const { ctx, parent } = await setup({ load: false })
     const self: { revoke?: () => void } = {}
@@ -405,10 +508,29 @@ describe('dsh-tool-subagent-report', () => {
   it('keeps the namespace plugin shape and validates its default', () => {
     expect('default' in tool).toBe(false)
     expect(tool.name).toBe('tool-subagent-report')
-    expect(tool.inject).toEqual(['subagents', 'tools'])
-    expect(tool.Config({}).reportDelivery).toBe('quiet')
+    expect(tool.inject).toEqual(['subagents', 'tools', 'systemPrompt'])
+    // Waking is the default because a report that never wakes its parent
+    // cannot deliver a result to an agent that already parked.
+    expect(tool.Config({}).reportDelivery).toBe('wakeup')
     expect(() => tool.Config({ reportDelivery: 'shout' } as never)).toThrow()
   })
+
+  it('wakes the parent under the default configuration', async () => {
+    const { ctx, parent, adapter } = await setup({ config: {} })
+    const { child } = await startChild(ctx, parent)
+    const enqueues: string[] = []
+    ctx.on('agent/inbox/inserted', ({ agent, message }) => {
+      if (agent === parent) {
+        enqueues.push(agent.inbox.nextTurn.some(queued => queued.id === message.id) ? 'queued' : 'steering')
+      }
+    })
+
+    expect((await callReport(ctx, child, 'DEFAULT_WAKES')).isError).toBe(false)
+    expect(enqueues).toEqual(['queued'])
+    await vi.waitFor(() => {
+      expect(adapter.requests.some(request => request.sessionId === parent.id)).toBe(true)
+    })
+  })
 })
 
 /** Prove report delivery uses ordinary logged user messages (runtime-context snapshots excluded). */
@@ -427,6 +549,9 @@ describe('dsh-tool-subagent-report result independence', () => {
       expect(ctx.agents.get(started.childId) === undefined).toBe(true)
     }, { timeout: 5_000 })
 
+    // The parent does learn the child settled — that account is the
+    // continuation service's, carried under its own `subagent-settled` source.
+    // Nothing turns the child's final answer into a report it did not send.
     expect(reports(parent)).toEqual([])
     expect(userTexts((await ctx.sessionPersistence.load(started.childId)).events)).toEqual(['child task'])
     expect(ctx.get('tasks')).toBeUndefined()
diff --git a/packages/subagent/tool-subagent-report/tsconfig.json b/packages/subagent/tool-subagent-report/tsconfig.json
index 9ccaad69f0..1a2a08671a 100644
--- a/packages/subagent/tool-subagent-report/tsconfig.json
+++ b/packages/subagent/tool-subagent-report/tsconfig.json
@@ -17,6 +17,9 @@
     {
       "path": "../../llm/llm"
     },
+    {
+      "path": "../../core/system-prompt"
+    },
     {
       "path": "../../core/tools"
     },
diff --git a/packages/subagent/tool-subagent/README.i18n.yaml b/packages/subagent/tool-subagent/README.i18n.yaml
index 7e038e3fa7..9cd884c016 100644
--- a/packages/subagent/tool-subagent/README.i18n.yaml
+++ b/packages/subagent/tool-subagent/README.i18n.yaml
@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write packages/subagent/tool-subagent/README.md
-README.md: 578ea4786e1d996251360c4aed92f2e882553681
-README.zh.md: baf91664530a637df96f1e7a51e6f0adabc0a8ae
+README.md: d4c472d54f87d6006aef96079050cf14885ed8fb
+README.zh.md: 36f79ac879067dc74fcc6b96122088bb68d53321
diff --git a/packages/subagent/tool-subagent/README.md b/packages/subagent/tool-subagent/README.md
index 578ea4786e..d4c472d54f 100644
--- a/packages/subagent/tool-subagent/README.md
+++ b/packages/subagent/tool-subagent/README.md
@@ -10,7 +10,7 @@ Each plugin instance binds one `provider` to one `toolName`; the model receives
 
 A foreground call passes the execution signal through startup and execution, awaits `run.result`, and always awaits `run.dispose()` before returning. Only `completed` returns the canonical `{ kind: 'foreground', runId, output: JsonValue[] }`, rendered as the same final text; abort, refusal, token limit, and other failures become errored tool results whose message appends the child's preserved partial text (the `SubagentResult.output` selection) after the stop-reason headline, so a truncated answer is never reported as success yet never silently lost. If result collection and disposal both reject, the errored result preserves both diagnostics.
 
-With `run_in_background: true`, `backgroundMode` selects the route. `one-shot` registers a plain parent-owned Task and returns canonical `{ kind: 'background', taskId }`, rendered as `started background subagent task <id>`, even when the provider supports continuable children; generic task tools own its later status, collection, cancellation, and notices. `continuable` requires a provider with the `prepareContinuable` capability, calls `ctx.subagents.startContinuable()`, and returns `{ kind: 'continuable', subagentId }`, rendered as `started subagent <childId>`. The continuable route resolves at inbox acceptance: the child owns its own turns from there, so this call neither waits for nor collects a result, and the child does not report back — its transcript by that id is the source of its output, and the optional global `send_message` tool sends it more work. Starting continuable work does not require `send_message` to be loaded. See the [background subagent Agent Note](../../../.agents/notes/implemented/feature/2026-07-08-background-subagent-tasks.md), the [continuable subagents Agent Note](../../../.agents/notes/implemented/feature/2026-07-28-continuable-subagent-conversations.md), and the [merged-service Agent Note](../../../.agents/notes/implemented/simplification/2026-07-26-merge-subagent-control-service.md).
+With `run_in_background: true`, `backgroundMode` selects the route. `one-shot` registers a plain parent-owned Task and returns canonical `{ kind: 'background', taskId }`, rendered as `started background subagent task <id>`, even when the provider supports continuable children; generic task tools own its later status, collection, cancellation, and notices. `continuable` requires a provider with the `prepareContinuable` capability, calls `ctx.subagents.startContinuable()`, and returns `{ kind: 'continuable', subagentId }`, rendered as `started subagent <childId>`. The continuable route resolves at inbox acceptance: the child owns its own turns from there, so this call neither waits for nor collects a result. The child's transcript by that id remains the source of its detailed output, and the optional global `send_message` tool sends it more work. The parent is not left guessing when to look, though: the continuation service delivers one settlement notice to it whenever a continuable child's Activation ends, which is why the schema tells the model it will be told and must not poll. Starting continuable work does not require `send_message` to be loaded. See the [background subagent Agent Note](../../../.agents/notes/implemented/feature/2026-07-08-background-subagent-tasks.md), the [continuable subagents Agent Note](../../../.agents/notes/implemented/feature/2026-07-28-continuable-subagent-conversations.md), and the [merged-service Agent Note](../../../.agents/notes/implemented/simplification/2026-07-26-merge-subagent-control-service.md).
 
 `toolFilter` changes the child's global tool layer but is not a parent-derived authority ceiling. See the [agent-scope security non-goal](../../../.agents/notes/implemented/architecture/2026-07-08-agent-scope-contexts.md#security-and-authority-are-non-goals).
 
@@ -37,7 +37,7 @@ Foreground and background calls are concurrency-safe: sibling delegations in one
 
 #### What the model sees
 
-The generated default [`subagent` schema](../../../docs/tool-catalog.md#deepseek-aidsh-tool-subagent) under this instance's configured name while its provider exists. Provider context inheritance changes the tool and prompt descriptions; enabled background mode adds `run_in_background`, and continuable mode describes starting a background subagent that keeps its conversation and returns its subagent id, while one-shot mode describes a background task id collected with `task_output` and stopped with `task_kill`.
+The generated default [`subagent` schema](../../../docs/tool-catalog.md#deepseek-aidsh-tool-subagent) under this instance's configured name while its provider exists. Provider context inheritance changes the tool and prompt descriptions; enabled background mode adds `run_in_background`, and continuable mode describes starting a background subagent that keeps its conversation, returns its subagent id, and reports its own completion — so the model is told never to poll or wait on it — while one-shot mode describes a background task id collected with `task_output` and stopped with `task_kill`.
 
 #### Token effect
 
@@ -65,11 +65,11 @@ Append-only; newly visible content follows the reusable request prefix and does
 
 #### What the model sees
 
-Start returns exactly `started subagent <childId>` in configured continuable mode, or `started background subagent task <id>` in configured one-shot mode. In one-shot mode the generic task surface provides later status, final output, cancellation responses, and notices. In continuable mode the child does not report back; an independently loaded `send_message` tool delivers follow-ups, and the child's transcript by its id is the source of its output.
+Start returns exactly `started subagent <childId>` in configured continuable mode, or `started background subagent task <id>` in configured one-shot mode. In one-shot mode the generic task surface provides later status, final output, cancellation responses, and notices. In continuable mode this tool returns no result of its own; the child's settlement reaches the parent as a [service-owned notice](../subagent/README.md#settlement-notice), an independently loaded `send_message` tool delivers follow-ups, and the child's transcript by its id is the source of its detailed output.
 
 #### Token effect
 
-The acknowledgement is retained; a one-shot final output enters parent history only when collected or injected, while a continuable child's output never returns through this tool.
+The acknowledgement is retained; a one-shot final output enters parent history only when collected or injected, while a continuable child's output never returns through this tool — its settlement notice arrives independently of any tool result.
 
 #### KV Cache effect
 
@@ -77,6 +77,6 @@ Append-only; newly visible content follows the reusable request prefix and does
 
 ## Known Limitations and Deferred Work
 
-- **Background runs expose no result through this tool** — a one-shot task's final output is collected through the generic task surface, and a continuable child's output stays in its own session, read by its subagent id.
+- **Background runs expose no result through this tool** — a one-shot task's final output is collected through the generic task surface, and a continuable child's output stays in its own session, read by its subagent id. The settlement notice states how that child ended and carries its closing message, but it is not this call's return value and cannot be awaited here.
 - **Duplicate names across waiting instances are detected late** (`TODO(subagent-dup-toolname)`) — preventing provider-registration rollback requires a registry of intended names.
 - **Child policy is fixed per instance** — another model, persona, tool filter, or depth cap requires another distinctly named tool.
diff --git a/packages/subagent/tool-subagent/README.zh.md b/packages/subagent/tool-subagent/README.zh.md
index baf9166453..36f79ac879 100644
--- a/packages/subagent/tool-subagent/README.zh.md
+++ b/packages/subagent/tool-subagent/README.zh.md
@@ -10,7 +10,7 @@
 
 前台调用会让执行信号贯穿启动和执行，等待 `run.result`，并且在返回前总会等待 `run.dispose()`。只有 `completed` 会返回规范值 `{ kind: 'foreground', runId, output: JsonValue[] }`，并渲染为相同的最终文本；中止、拒绝、token 上限和其他失败都会变成出错的工具结果，其消息在终止原因标题之后附带子代理保留下来的部分文本（即 `SubagentResult.output` 的选取结果）——被截断的回答不会被报告为成功，也绝不会被悄悄丢弃。如果结果收集与 dispose（资源释放）都 reject，出错的结果会保留两项诊断信息。
 
-设置 `run_in_background: true` 后，`backgroundMode` 会选择路由。`one-shot` 会注册一个归父级所有的普通 Task，并返回规范值 `{ kind: 'background', taskId }`，渲染为 `started background subagent task <id>`，即使提供方支持可继续子 agent 也不例外；通用 Task 工具负责其后续状态、收集、取消和通知。`continuable` 要求提供方具备 `prepareContinuable` 能力，调用 `ctx.subagents.startContinuable()`，并返回 `{ kind: 'continuable', subagentId }`，渲染为 `started subagent <childId>`。可继续路由在 inbox 接受时结算：子 agent 自此拥有自己的轮次，因此该调用既不等待也不收集结果，而且子 agent 不会回报——通过该 id 查看其 transcript（文本记录）即是其输出来源，可选的全局 `send_message` 工具则向其发送更多工作。启动可继续工作不要求加载 `send_message`。见 [后台 subagent Agent Note](../../../.agents/notes/implemented/feature/2026-07-08-background-subagent-tasks.md)、[可继续的 subagent Agent Note](../../../.agents/notes/implemented/feature/2026-07-28-continuable-subagent-conversations.md)和[服务合并 Agent Note](../../../.agents/notes/implemented/simplification/2026-07-26-merge-subagent-control-service.md)。
+设置 `run_in_background: true` 后，`backgroundMode` 会选择路由。`one-shot` 会注册一个归父级所有的普通 Task，并返回规范值 `{ kind: 'background', taskId }`，渲染为 `started background subagent task <id>`，即使提供方支持可继续子 agent 也不例外；通用 Task 工具负责其后续状态、收集、取消和通知。`continuable` 要求提供方具备 `prepareContinuable` 能力，调用 `ctx.subagents.startContinuable()`，并返回 `{ kind: 'continuable', subagentId }`，渲染为 `started subagent <childId>`。可继续路由在 inbox 接受时结算：子 agent 自此拥有自己的轮次，因此该调用既不等待也不收集结果。通过该 id 查看其 transcript（文本记录）仍是其详细输出的来源，可选的全局 `send_message` 工具则向其发送更多工作。不过父级无需猜测何时查看：每当可继续子 agent 的 Activation 结束，继续执行服务都会向父级投递一条结算通知——正因如此，schema 才告诉模型它会被通知，且不得轮询。启动可继续工作不要求加载 `send_message`。见 [后台 subagent Agent Note](../../../.agents/notes/implemented/feature/2026-07-08-background-subagent-tasks.md)、[可继续的 subagent Agent Note](../../../.agents/notes/implemented/feature/2026-07-28-continuable-subagent-conversations.md)和[服务合并 Agent Note](../../../.agents/notes/implemented/simplification/2026-07-26-merge-subagent-control-service.md)。
 
 `toolFilter` 会改变子 agent 的全局工具层，但不是从父级派生的权限上限。见 [agent 作用域的安全非目标](../../../.agents/notes/implemented/architecture/2026-07-08-agent-scope-contexts.md#security-and-authority-are-non-goals)。
 
@@ -37,7 +37,7 @@
 
 #### 模型看到的内容
 
-当提供方存在时，以当前实例配置的名称公开已生成的默认 [`subagent` schema](../../../docs/tool-catalog.md#deepseek-aidsh-tool-subagent)。提供方是否继承上下文会改变工具描述和提示词描述；启用后台模式会添加 `run_in_background`，可继续模式描述为启动一个保留其对话并返回子 agent id 的后台子 agent，而一次性模式描述为返回一个用 `task_output` 收集、用 `task_kill` 停止的后台任务 id。
+当提供方存在时，以当前实例配置的名称公开已生成的默认 [`subagent` schema](../../../docs/tool-catalog.md#deepseek-aidsh-tool-subagent)。提供方是否继承上下文会改变工具描述和提示词描述；启用后台模式会添加 `run_in_background`，可继续模式描述为启动一个保留其对话、返回子 agent id 并会自行报告完成的后台子 agent——因此模型被告知绝不要轮询或等待它——而一次性模式描述为返回一个用 `task_output` 收集、用 `task_kill` 停止的后台任务 id。
 
 #### Token 影响
 
@@ -65,11 +65,11 @@
 
 #### 模型看到的内容
 
-在配置的可继续模式下，启动时返回内容恰为 `started subagent <childId>`；在配置的一次性模式下，则返回 `started background subagent task <id>`。一次性模式下，通用 Task 接口提供后续状态、最终输出、取消响应和通知。可继续模式下，子 agent 不会回报；独立加载的 `send_message` 工具会投递后续消息，而通过其 id 查看子 agent 的 transcript 即是其输出来源。
+在配置的可继续模式下，启动时返回内容恰为 `started subagent <childId>`；在配置的一次性模式下，则返回 `started background subagent task <id>`。一次性模式下，通用 Task 接口提供后续状态、最终输出、取消响应和通知。可继续模式下，本工具不返回自己的结果；子 agent 的结算会以[服务负责的通知](../subagent/README.md#settlement-notice)到达父级，独立加载的 `send_message` 工具会投递后续消息，而通过其 id 查看子 agent 的 transcript 即是其详细输出来源。
 
 #### Token 影响
 
-确认消息会被保留；一次性最终输出只在收集或注入时进入父级历史，而可继续子 agent 的输出绝不会通过本工具返回。
+确认消息会被保留；一次性最终输出只在收集或注入时进入父级历史，而可继续子 agent 的输出绝不会通过本工具返回——其结算通知独立于任何工具结果到达。
 
 #### KV Cache 影响
 
@@ -77,6 +77,6 @@
 
 ## 已知限制与暂缓事项
 
-- **后台运行不通过本工具公开结果**：一次性任务的最终输出通过通用 Task 接口收集，可继续子 agent 的输出留在其自身会话中，按其 subagent id 读取。
+- **后台运行不通过本工具公开结果**：一次性任务的最终输出通过通用 Task 接口收集，可继续子 agent 的输出留在其自身会话中，按其 subagent id 读取。结算通知会说明该子 agent 如何结束并携带其收尾消息，但它不是本次调用的返回值，也无法在此等待。
 - **等待中实例的重复名称发现较晚**（`TODO(subagent-dup-toolname)`）：若要阻止提供方注册回滚，需要一份预期名称注册表。
 - **每个实例的子 agent 策略固定**：其他模型、persona、工具过滤器或深度上限都需要另一个名称不同的工具。
diff --git a/packages/subagent/tool-subagent/src/index.ts b/packages/subagent/tool-subagent/src/index.ts
index 88ed0cdfc1..b84ee28da9 100644
--- a/packages/subagent/tool-subagent/src/index.ts
+++ b/packages/subagent/tool-subagent/src/index.ts
@@ -262,12 +262,13 @@ export function apply(ctx: Context, config: Config): void {
     disposeTool = ctx.tools.register(defineTool({
       name: config.toolName ?? 'subagent',
       description: wording.description + (backgroundEnabled
-        // The return channel is a separately installed capability this package
-        // cannot observe, so this describes only this call's result.
+        // The completion notice is the continuation service's own behavior, not
+        // a separately installed capability, so this promise holds whenever the
+        // continuable background path is reachable at all.
         ? continuable
           ? ' Set `run_in_background: true` to start a background subagent that keeps its conversation:'
-          + ' you receive only its subagent id, never its result, and it works on its own. Use this for'
-          + ' work whose result you do not need returned by this call; `send_message` sends it more work.'
+          + ' this call returns only its subagent id, and the subagent works on its own from there. You'
+          + ' are told when it finishes, so never poll or wait on it; `send_message` sends it more work.'
           : ' Set `run_in_background: true` to return a task id; collect with `task_output` and stop with `task_kill`.'
         : ''),
       parameters: {
@@ -286,7 +287,7 @@ export function apply(ctx: Context, config: Config): void {
             type: 'boolean' as const,
             description: continuable
               ? 'Run as a background subagent that keeps its conversation and return only its subagent id. '
-              + 'This call never returns its result; send it more work with send_message.'
+              + 'This call does not wait for it; you are told when it finishes. Send it more work with send_message.'
               : 'Run as a background task and return its id; collect with task_output or stop with task_kill.',
           },
         } : {},
diff --git a/packages/support/acp-snapshot/README.i18n.yaml b/packages/support/acp-snapshot/README.i18n.yaml
index 5898c5beb1..ca9e9e67f0 100644
--- a/packages/support/acp-snapshot/README.i18n.yaml
+++ b/packages/support/acp-snapshot/README.i18n.yaml
@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write packages/support/acp-snapshot/README.md
-README.md: 8a9669b1f555fceb7af6d017d273b62d7dcad1e1
-README.zh.md: efa31df86d3b494ab14fc3742fc84adce570a593
+README.md: 06f1cb67cfcd954254db480ea696d10d81b37438
+README.zh.md: c4a5643f8c5d7e5b62a160cd32e5969187d34046
diff --git a/packages/support/acp-snapshot/README.md b/packages/support/acp-snapshot/README.md
index 8a9669b1f5..06f1cb67cf 100644
--- a/packages/support/acp-snapshot/README.md
+++ b/packages/support/acp-snapshot/README.md
@@ -55,6 +55,8 @@ A scenario booting a differently-composed tree sets its own `configPath` (an ove
 
 A pin owns its generated `system-prompt.expected.md` or `tool-schemas.expected.json` by default; `systemPromptSource` and `toolSchemasSource` name another pin when the complete corresponding sequence is identical, so each distinct version is committed once. The pin's `session.jsonl` stores `"system":"{{system}}","tools":"{{tools}}"` while retaining config, reason, and any model-visible prefix. A pin with legitimate mid-run header changes declares `expectedHeaderChanges`; a shared source must declare the same count, and record/refresh rejects claimants that generate different bytes.
 
+A child session whose own scope composes a different request declares it per fixture index: `pinsChildToolSchemas` moves that child's tool sequence into `tool-schemas.<n>.expected.json`, and `pinsChildSystemPrompts` moves its prompt into `system-prompt.<n>.expected.md`. Each names the `session.<n>.jsonl` fixture it describes, leaves every other request-header field to the class pin, and requires its sidecar to exist exactly when declared. A child prompt sidecar must also differ from its class pin, so a redundant copy fails instead of drifting. A continuable child carrying the scope-local `report` tool and its guidance section is the shipped case for both.
+
 Every scenario compares `stdout.expected.jsonl` with cwd-rooted separators canonicalized to `/`. On Windows, `pinsNativeWindowsStdout` additionally compares the complete `stdout.expected.windows.jsonl` after the shared expected output and requires that sidecar exactly when enabled. A scenario requiring a non-Windows host declares `posixOnly`, which skips its run test on Windows while the fixture guards keep covering its committed files everywhere; examples include POSIX process semantics (e.g. cancelling a live bash call kills a detached process group) and generated paths Windows cannot represent. A scenario whose composition needs a usable `pwsh` declares `pwshOnly`; the caller-supplied `hasPwsh` probe (the shipped acp-agent suite follows the executor's own resolution, so Program Files installs count) skips the run test when no usable `pwsh` resolves while the fixture guards keep covering its committed files everywhere.
 
 The example also ships a `cordis.snapshot.yml` replay overlay next to its `cordis.yml` (the bin swaps them under `DSH_SNAPSHOT=replay` — [single-source replay config Agent Note](../../../.agents/notes/archived/testing/2026-07-04-single-source-acp-replay-config.md)); replay fixtures are served by [`dsh-llm-replay`](../llm-replay/README.md), which this package points at via the `DSH_SNAPSHOT_*` env vars it sets on the child. `pnpm run test:snapshot:record` calls the live LLM and rewrites the recorded scenarios' model fixtures; `pnpm run test:snapshot:refresh` stays keyless, runs the replay overlay, and rewrites stdout, comparable session-log expected outputs, and owned prompt and tool-schema sidecars from the committed model scripts. Fixture roles, record/replay/refresh semantics, and scenario-table fields are documented on `Scenario` and in the [snapshot Agent Note](../../../.agents/notes/implemented/testing/2026-06-19-acp-snapshot-tests.md).
diff --git a/packages/support/acp-snapshot/README.zh.md b/packages/support/acp-snapshot/README.zh.md
index efa31df86d..c4a5643f8c 100644
--- a/packages/support/acp-snapshot/README.zh.md
+++ b/packages/support/acp-snapshot/README.zh.md
@@ -55,6 +55,8 @@ defineAcpSnapshotSuite({
 
 每个 pin 默认拥有其生成的 `system-prompt.expected.md` 或 `tool-schemas.expected.json`；当完整的对应序列相同时，`systemPromptSource` 和 `toolSchemasSource` 指定另一个 pin 作为来源，因此每个不同版本只提交一次。该 pin 的 `session.jsonl` 存储 `"system":"{{system}}","tools":"{{tools}}"`，同时保留配置、原因和任何模型可见前缀。具有合法运行中 header 变更的 pin 声明 `expectedHeaderChanges`；共享来源必须声明相同的 header 变更数量，录制/刷新会拒绝生成不同字节的共享引用方。
 
+自身作用域组合出不同请求的 child 会话按 fixture 索引单独声明：`pinsChildToolSchemas` 把该 child 的工具序列移入 `tool-schemas.<n>.expected.json`，`pinsChildSystemPrompts` 把其提示词移入 `system-prompt.<n>.expected.md`。两者都指名自己描述的 `session.<n>.jsonl` fixture，其余请求 header 字段仍归类别 pin 所有，并要求 sidecar 恰好在声明时存在。child 提示词 sidecar 还必须与其类别 pin 不同，因此冗余副本会直接失败，而不会悄悄漂移。携带作用域局部 `report` 工具及其指引 section 的可继续 child 是两者的随附用例。
+
 每个场景都比较 `stdout.expected.jsonl`，其中以 cwd 为根的分隔符规范化为 `/`。在 Windows 上，`pinsNativeWindowsStdout` 还会在共享预期输出之后比较完整 `stdout.expected.windows.jsonl`，并在启用时精确要求该 sidecar。需要非 Windows 主机的场景声明 `posixOnly`，在 Windows 上跳过运行测试，但 fixture 保护仍在所有平台覆盖其已提交文件；示例包括 POSIX 进程语义（例如取消实时 bash 调用会终止脱离进程组）和 Windows 无法表示的生成路径。组合需要可用 `pwsh` 的场景声明 `pwshOnly`；调用方提供的 `hasPwsh` 探测（随附的 acp-agent 套件遵循执行器自身的解析，因此 Program Files 安装也计入）在解析不到可用 `pwsh` 时跳过运行测试，而 fixture 保护仍处处覆盖其已提交文件。
 
 示例还发布 `cordis.snapshot.yml` 回放 overlay，位于 `cordis.yml` 旁边（bin 在 `DSH_SNAPSHOT=replay` 下交换它们，见[单源回放配置 Agent Note](../../../.agents/notes/archived/testing/2026-07-04-single-source-acp-replay-config.md)）；回放 fixture 由 [`dsh-llm-replay`](../llm-replay/README.md) 提供，该包通过对子级设置的 `DSH_SNAPSHOT_*` env var 指向它。`pnpm run test:snapshot:record` 调用实时 LLM，并重写已记录场景的模型 fixture；`pnpm run test:snapshot:refresh` 保持无密钥，运行回放 overlay，并从已提交模型脚本重写 stdout、可比较会话日志预期输出，以及各 pin 自有的提示词与工具 schema sidecar。Fixture 角色、录制/回放/刷新语义和场景表字段记录在 `Scenario` 以及[快照 Agent Note](../../../.agents/notes/implemented/testing/2026-06-19-acp-snapshot-tests.md) 中。
diff --git a/packages/support/acp-snapshot/src/suite.ts b/packages/support/acp-snapshot/src/suite.ts
index 5cdffadf38..1728671c3a 100644
--- a/packages/support/acp-snapshot/src/suite.ts
+++ b/packages/support/acp-snapshot/src/suite.ts
@@ -46,6 +46,11 @@ function childToolSchemasSnapshot(index: number): string {
   return `tool-schemas.${index}.expected.json`
 }
 
+/** Return the dedicated system-prompt sidecar for one child fixture index. */
+function childSystemPromptSnapshot(index: number): string {
+  return `system-prompt.${index}.expected.md`
+}
+
 /** The optional full Windows-native stdout transcript. */
 const WINDOWS_STDOUT_SNAPSHOT = 'stdout.expected.windows.jsonl'
 
@@ -116,6 +121,13 @@ export interface Scenario {
    * request-header field.
    */
   pinsChildToolSchemas?: readonly number[]
+  /**
+   * Child fixture indices whose own system prompt is pinned separately, where
+   * `1` names `session.1.jsonl` and `system-prompt.1.expected.md`. A child
+   * scope that installs its own prompt section (the continuable `report`
+   * guidance) composes a prompt the class pin cannot describe.
+   */
+  pinsChildSystemPrompts?: readonly number[]
   /**
    * How many changed `request/header` snapshots this PINNING scenario's primary
    * fixture legitimately carries (default 0). Their full prompt text is kept in
@@ -491,6 +503,18 @@ export function formatSystemPromptSnapshot(
   return snapshot
 }
 
+/**
+ * Reject a child prompt sidecar that cannot own distinct, canonical prompt text.
+ * @param sidecar - committed child prompt snapshot.
+ * @param classPin - initial prompt snapshot owned by the scenario's header class.
+ * @param label - repository-relative fixture label for diagnostics.
+ */
+export function assertChildSystemPromptSnapshot(sidecar: string, classPin: string, label: string): void {
+  if (sidecar.trim().length === 0) throw new Error(`${label} must pin a non-empty prompt`)
+  if (!sidecar.endsWith('\n')) throw new Error(`${label} must end in a newline`)
+  if (sidecar === classPin) throw new Error(`${label} must differ from its class pin`)
+}
+
 /** Return the initial-prompt portion of a possibly multi-header snapshot. */
 function initialSystemPromptSnapshot(snapshot: string): string {
   const marker = snapshot.indexOf('\n<!-- request/header change ')
@@ -1192,6 +1216,7 @@ export function defineAcpSnapshotSuite(options: SnapshotSuiteOptions): void {
         }
 
         const childSchemaPins = new Set(scenario.pinsChildToolSchemas ?? [])
+        const childPromptPins = new Set(scenario.pinsChildSystemPrompts ?? [])
 
         // Record writes live model fixtures; keyless refresh writes every comparable replayed
         // fixture. Pinning JSONL keeps prefixes but moves prompts and schemas into sidecars.
@@ -1281,6 +1306,18 @@ export function defineAcpSnapshotSuite(options: SnapshotSuiteOptions): void {
               schemaSets.slice(1),
             ))
           }
+          for (const index of childPromptPins) {
+            const log = result.sessionLogs[index]
+            expect(log, `${mode}: no child session log at index ${index} to snapshot a prompt from`)
+              .toBeDefined()
+            const prompts = normalizedSystemPrompts((log as HarvestedLog).content, ctx)
+            expect(prompts.length, `${mode}: child ${index} produced no system prompt to snapshot`)
+              .toBeGreaterThan(0)
+            await writeFile(
+              join(dir, childSystemPromptSnapshot(index)),
+              formatSystemPromptSnapshot(prompts[0] as string),
+            )
+          }
         }
 
         for (const expected of stdoutExpectedVariants(scenario)) {
@@ -1340,6 +1377,13 @@ export function defineAcpSnapshotSuite(options: SnapshotSuiteOptions): void {
           const parsed = parseToolSchemasSnapshot(sidecar)
           childPinnedSchemas.set(index, [parsed.initial, ...parsed.changes])
         }
+        const childPinnedPrompts = new Map<number, string>()
+        for (const index of childPromptPins) {
+          childPinnedPrompts.set(
+            index,
+            await readFile(join(dir, childSystemPromptSnapshot(index)), 'utf8'),
+          )
+        }
         for (const [logIndex, log] of result.sessionLogs.entries()) {
           const childSchemas = childPinnedSchemas.get(logIndex)
           const expectedChanges = scenario.pinsHeader === true && logIndex === 0
@@ -1366,8 +1410,14 @@ export function defineAcpSnapshotSuite(options: SnapshotSuiteOptions): void {
             expect(header, `session ${log.id}: request/header #${k + 1} diverged from the pinned (${pinningScenario.name}) header`)
               .toEqual(expected)
             if (expectedChanges === 0) {
-              expect(formatSystemPromptSnapshot(prompts[k] as string), `session ${log.id}: initial system prompt #${k + 1} diverged from ${promptSource.name}/${SYSTEM_PROMPT_SNAPSHOT}`)
-                .toEqual(initialPromptSnapshot)
+              // A pinned child owns its whole prompt: its scope-local sections
+              // are exactly what the class pin cannot describe.
+              const childPrompt = childPinnedPrompts.get(logIndex)
+              const promptOrigin = childPrompt === undefined
+                ? `${promptSource.name}/${SYSTEM_PROMPT_SNAPSHOT}`
+                : childSystemPromptSnapshot(logIndex)
+              expect(formatSystemPromptSnapshot(prompts[k] as string), `session ${log.id}: initial system prompt #${k + 1} diverged from ${promptOrigin}`)
+                .toEqual(childPrompt ?? initialPromptSnapshot)
             }
           }
           if (scenario.pinsHeader === true && logIndex === 0) {
@@ -1400,16 +1450,19 @@ export function defineAcpSnapshotSuite(options: SnapshotSuiteOptions): void {
 
     it('every registered scenario has its required fixture files', async () => {
       // Every scenario needs input, stdout, a primary session fixture, and matching optional sidecars.
-      for (const { name, overridden, pinsNativeWindowsStdout, pinsChildToolSchemas } of scenarios) {
+      for (const { name, overridden, pinsNativeWindowsStdout, pinsChildToolSchemas, pinsChildSystemPrompts } of scenarios) {
         const dir = join(snapshotsDir, name)
-        const declaredChildPins = new Set(pinsChildToolSchemas ?? [])
-        const childSidecars = (await readdir(dir, { withFileTypes: true }))
+        const files = (await readdir(dir, { withFileTypes: true }))
           .filter(entry => entry.isFile())
-          .map(entry => /^tool-schemas\.([1-9]\d*)\.expected\.json$/.exec(entry.name))
+          .map(entry => entry.name)
+        const childIndices = (pattern: RegExp): Set<number> => new Set(files
+          .map(file => pattern.exec(file))
           .filter((match): match is RegExpExecArray => match !== null)
-          .map(match => Number(match[1]))
-        expect(new Set(childSidecars), `${name}: child tool-schema sidecars must match \`pinsChildToolSchemas\``)
-          .toEqual(declaredChildPins)
+          .map(match => Number(match[1])))
+        expect(childIndices(/^tool-schemas\.([1-9]\d*)\.expected\.json$/), `${name}: child tool-schema sidecars must match \`pinsChildToolSchemas\``)
+          .toEqual(new Set(pinsChildToolSchemas ?? []))
+        expect(childIndices(/^system-prompt\.([1-9]\d*)\.expected\.md$/), `${name}: child system-prompt sidecars must match \`pinsChildSystemPrompts\``)
+          .toEqual(new Set(pinsChildSystemPrompts ?? []))
         expect(existsSync(join(dir, 'input.json')), `${name}/input.json`).toBe(true)
         expect(existsSync(join(dir, 'stdout.expected.jsonl')), `${name}/stdout.expected.jsonl`).toBe(true)
         expect(
@@ -1492,13 +1545,11 @@ export function defineAcpSnapshotSuite(options: SnapshotSuiteOptions): void {
       assertUniqueSnapshotContents('tool-schema', schemas)
     })
 
-    it('every declared child tool-schema sidecar is canonical and names a real child', async () => {
+    it('every declared child sidecar is canonical and names a real child', async () => {
       for (const scenario of scenarios) {
-        const pins = scenario.pinsChildToolSchemas ?? []
-        if (pins.length === 0) continue
         const dir = join(snapshotsDir, scenario.name)
         const files = await sessionFixtures(dir)
-        for (const index of pins) {
+        for (const index of scenario.pinsChildToolSchemas ?? []) {
           expect(files[index], `${scenario.name}: child schema pin ${index} must name an existing session.<n>.jsonl fixture`)
             .toBeDefined()
           const file = childToolSchemasSnapshot(index)
@@ -1509,6 +1560,16 @@ export function defineAcpSnapshotSuite(options: SnapshotSuiteOptions): void {
           expect(parsed.initial.length, `${scenario.name}/${file} must pin at least one schema`)
             .toBeGreaterThan(0)
         }
+        for (const index of scenario.pinsChildSystemPrompts ?? []) {
+          expect(files[index], `${scenario.name}: child prompt pin ${index} must name an existing session.<n>.jsonl fixture`)
+            .toBeDefined()
+          const file = childSystemPromptSnapshot(index)
+          const sidecar = await readFile(join(dir, file), 'utf8')
+          /* v8 ignore next -- registration guarantees every scenario class has resolved sources. */
+          const promptSource = promptSourceByClass.get(classOf(scenario)) ?? scenario
+          const classPin = await readFile(join(snapshotsDir, promptSource.name, SYSTEM_PROMPT_SNAPSHOT), 'utf8')
+          assertChildSystemPromptSnapshot(sidecar, initialSystemPromptSnapshot(classPin), `${scenario.name}/${file}`)
+        }
       }
     })
 
diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/behavior.json b/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/behavior.json
index 8599498bee..b4e3667519 100644
--- a/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/behavior.json
+++ b/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/behavior.json
@@ -9,7 +9,7 @@
     ]},
     { "file": "b/child/session.jsonl", "lines": [
       { "type": "session", "id": "eeeeeeee-1111-4222-8333-444444444444", "createdAt": 300, "cwd": "{{CWD}}", "parentSession": "{{SID}}", "delegationDepth": 1 },
-      { "type": "request/header", "seq": 0, "time": 6, "data": { "header": { "config": { "model": "fake" }, "system": "SYS PROMPT", "tools": [{ "name": "child-only", "description": "Child D", "parameters": { "type": "object" } }] }, "reason": "initial" } }
+      { "type": "request/header", "seq": 0, "time": 6, "data": { "header": { "config": { "model": "fake" }, "system": "SYS PROMPT\n\nCHILD GUIDANCE", "tools": [{ "name": "child-only", "description": "Child D", "parameters": { "type": "object" } }] }, "reason": "initial" } }
     ]}
   ]
 }
diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/system-prompt.1.expected.md b/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/system-prompt.1.expected.md
new file mode 100644
index 0000000000..56f56d3a6a
--- /dev/null
+++ b/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/system-prompt.1.expected.md
@@ -0,0 +1,3 @@
+SYS PROMPT
+
+CHILD GUIDANCE
diff --git a/packages/support/acp-snapshot/tests/suite.spec.ts b/packages/support/acp-snapshot/tests/suite.spec.ts
index 3cfd1c0f0a..8f072bc2e5 100644
--- a/packages/support/acp-snapshot/tests/suite.spec.ts
+++ b/packages/support/acp-snapshot/tests/suite.spec.ts
@@ -12,6 +12,7 @@ import {
   type Scenario,
 } from '../src/index.ts'
 import {
+  assertChildSystemPromptSnapshot,
   assertUniqueSnapshotContents,
   claimSharedSnapshot,
   fixtureContext,
@@ -85,6 +86,7 @@ const REPLAY_SCENARIOS: Scenario[] = [
     configPath: AGENT.configPath,
     workspaceParent: tmpdir(),
     pinsChildToolSchemas: [1],
+    pinsChildSystemPrompts: [1],
     prepareWorkspace: (cwd) => {
       writeFileSync(join(cwd, 'seed.txt'), 'prepared at runtime')
     },
@@ -126,6 +128,7 @@ function staleRefreshFixtures(dir: string): void {
   writeFileSync(join(dir, 'pin-turn', 'system-prompt.expected.md'), 'STALE PROMPT\n')
   writeFileSync(join(dir, 'pin-turn', 'tool-schemas.expected.json'), '{"initial":[{"name":"stale"}],"changes":[]}\n')
   writeFileSync(join(dir, 'plain-turn', 'tool-schemas.1.expected.json'), '{"initial":[{"name":"stale-child"}],"changes":[]}\n')
+  writeFileSync(join(dir, 'plain-turn', 'system-prompt.1.expected.md'), 'STALE CHILD PROMPT\n')
 
   const plainBehaviorFile = join(dir, 'plain-turn', 'behavior.json')
   const plainBehavior = JSON.parse(readFileSync(plainBehaviorFile, 'utf8')) as Record<string, unknown>
@@ -191,6 +194,8 @@ describe('defineAcpSnapshotSuite: refresh write-back', () => {
     const childSchemas = readFileSync(join(refreshDir, 'plain-turn', 'tool-schemas.1.expected.json'), 'utf8')
     expect(childSchemas).toContain('"name": "child-only"')
     expect(childSchemas).not.toContain('stale-child')
+    const childPrompt = readFileSync(join(refreshDir, 'plain-turn', 'system-prompt.1.expected.md'), 'utf8')
+    expect(childPrompt).toBe('SYS PROMPT\n\nCHILD GUIDANCE\n')
 
     const pinSession = readFileSync(join(refreshDir, 'pin-turn', 'session.jsonl'), 'utf8')
     expect(pinSession).toContain('"cwd":"{{cwd}}"')
@@ -591,6 +596,34 @@ describe('formatSystemPromptSnapshot', () => {
   })
 })
 
+describe('assertChildSystemPromptSnapshot', () => {
+  const label = 'plain-turn/system-prompt.1.expected.md'
+
+  it('accepts a distinct non-empty canonical child prompt', () => {
+    expect(() => {
+      assertChildSystemPromptSnapshot('SYS PROMPT\n\nCHILD GUIDANCE\n', 'SYS PROMPT\n', label)
+    }).not.toThrow()
+  })
+
+  it('rejects an empty or non-canonical child prompt', () => {
+    expect(() => { assertChildSystemPromptSnapshot('\n', 'SYS PROMPT\n', label) }).toThrow(/non-empty prompt/)
+    expect(() => {
+      assertChildSystemPromptSnapshot('CHILD GUIDANCE', 'SYS PROMPT\n', label)
+    }).toThrow(/end in a newline/)
+  })
+
+  it('rejects a child prompt that duplicates its class pin', () => {
+    const classSnapshot = readFileSync(join(REPLAY_DIR, 'pin-turn', 'system-prompt.expected.md'), 'utf8')
+    const marker = classSnapshot.indexOf('\n<!-- request/header change ')
+    expect(marker).toBeGreaterThan(0)
+    const initialClassPin = classSnapshot.slice(0, marker)
+
+    expect(() => {
+      assertChildSystemPromptSnapshot(initialClassPin, initialClassPin, label)
+    }).toThrow(/must differ from its class pin/)
+  })
+})
+
 describe('headerChangeCount', () => {
   it('counts changed request headers, ignoring anchors, blanks, and other lines', () => {
     const change = JSON.stringify({ type: 'request/header', seq: 2, time: 9, data: { reason: 'change' } })
diff --git a/pnpm-lock.yaml b/pnpm-lock.yaml
index e5e6120f47..d27a1bc6f2 100644
--- a/pnpm-lock.yaml
+++ b/pnpm-lock.yaml
@@ -6846,6 +6846,9 @@ importers:
       '@deepseek-ai/dsh-subagent-spawn':
         specifier: workspace:^
         version: link:../subagent-spawn
+      '@deepseek-ai/dsh-system-prompt':
+        specifier: workspace:^
+        version: link:../../core/system-prompt
       '@deepseek-ai/dsh-tool-subagent-control':
         specifier: workspace:^
         version: link:../tool-subagent-control
diff --git a/scripts/gen-tool-catalog.ts b/scripts/gen-tool-catalog.ts
index 76878426fe..a2ff4ce533 100644
--- a/scripts/gen-tool-catalog.ts
+++ b/scripts/gen-tool-catalog.ts
@@ -32,7 +32,7 @@ import WebService from '@deepseek-ai/dsh-web'
 import * as WebSearchExa from '@deepseek-ai/dsh-web-search-exa'
 import * as WebFetchLocal from '@deepseek-ai/dsh-web-fetch-local'
 import SubagentService from '@deepseek-ai/dsh-subagent'
-import type { SubagentProvider } from '@deepseek-ai/dsh-subagent'
+import type { SubagentProvider, SubagentReportDelivery } from '@deepseek-ai/dsh-subagent'
 import * as ToolSubagentControl from '@deepseek-ai/dsh-tool-subagent-control'
 import * as ToolSubagentListAgents from '@deepseek-ai/dsh-tool-subagent-control/list-agents'
 import * as ToolSubagentReport from '@deepseek-ai/dsh-tool-subagent-report'
@@ -420,7 +420,7 @@ const TOOL_PACKAGES: ToolPackage[] = [
       await ctx.plugin(ToolSubagent, { provider: 'mock' })
     },
     note:
-      'The registered tool name is the load-time `toolName` config (default `subagent`); the schema above is that default. The shipped example agents load this package once per subagent backend, so the model additionally sees `subagent_fork` (bound to the fork backend) with an identical schema — see `packages/bundle/base/cordis.patch.yml` and `examples/acp-agent/cordis.yml`.',
+      'The registered tool name is the load-time `toolName` config (default `subagent`); the schema above is that default. The shipped compositions load this package once per subagent backend, so the model additionally sees `subagent_fork` bound to the fork backend. Each instance\'s description and `run_in_background` parameter follow its own `backgroundMode` and `enableRunInBackground`, so the two shipped schemas are not identical: `subagent` is `continuable`, while `subagent_fork` stays `one-shot` — see `packages/bundle/base/cordis.patch.yml` and `examples/acp-agent/cordis.yml`.',
   },
   {
     pkg: '@deepseek-ai/dsh-tool-subagent-control',
@@ -448,20 +448,22 @@ const TOOL_PACKAGES: ToolPackage[] = [
     pkg: '@deepseek-ai/dsh-tool-subagent-report',
     dir: 'tool-subagent-report',
     source: 'packages/subagent/tool-subagent-report/src/index.ts',
-    requires: ['ctx.subagents', 'a live continuable in-process child Agent'],
+    requires: ['ctx.subagents', 'ctx.systemPrompt', 'a live continuable in-process child Agent'],
     writes: ['tool/call', 'tool/result', 'a user-role message in the direct parent session'],
     async mount(ctx) {
       await ctx.plugin(AgentRegistry)
       await ctx.plugin(SubagentService)
+      const { reportDelivery } = ToolSubagentReport.Config({}) as { reportDelivery: SubagentReportDelivery }
       await mountCatalogChildScope(ctx, (childCtx) => {
-        ToolSubagentReport.installReportTool(childCtx, ctx, 'quiet')
+        ToolSubagentReport.installReportTool(childCtx, ctx, reportDelivery)
       })
     },
     scope: ctx => catalogChildScopes.get(ctx) as Agent,
     note:
       'Registered per continuable in-process child rather than globally, so this schema is visible only '
-      + 'inside such a child and survives its global `toolFilter`. The parent-facing `send_message` tool '
-      + 'is installed independently.',
+      + 'inside such a child and survives its global `toolFilter`. The same contribution installs the '
+      + 'child-scoped `tool:report` prompt section, which this catalog does not render. The parent-facing '
+      + '`send_message` tool is installed independently.',
   },
   {
     pkg: '@deepseek-ai/dsh-tool-tasks',
diff --git a/scripts/type-equiv.manifest.json b/scripts/type-equiv.manifest.json
index 11de45ad78..86de2d6eb5 100644
--- a/scripts/type-equiv.manifest.json
+++ b/scripts/type-equiv.manifest.json
@@ -1170,6 +1170,11 @@
       "symbol": "SubagentReportDelivery",
       "source": "packages/subagent/subagent/src/continuation.ts"
     },
+    {
+      "doc": "docs/subsystems/subagent.md",
+      "symbol": "SubagentSettledMessageSource",
+      "source": "packages/subagent/subagent/src/continuation.ts"
+    },
     {
       "doc": "docs/subsystems/subagent.md",
       "symbol": "SubagentReportOptions",