mirror of
https://github.com/deepseek-ai/deepseek-harness
synced 2026-08-15 21:04:50 +00:00
fix(session-title): respect request and config boundaries
This commit is contained in:
Tianyi Cui
@@ -12,7 +12,7 @@ The reference shape for the happy path is MiniCode's `LLMClient`: a stateful con
|
||||
|
||||
### The principle
|
||||
|
||||
**Model-visible ⟺ logged.** Anything that reaches a model request must be recorded in the session log. The checkable consequence: **every conversation request the loop sends is a pure function of the session log** — anyone holding the log reconstructs it byte-for-byte. Scope, stated precisely: the guarantee covers the loop-built `GenerateOptions`; provider wire bytes follow from it because both adapters' serialization is a pure per-message function at a pinned code version; direct one-shots (compaction's summarize call) log their envelope scalars (`compact/summary.{provider, model, maxTokens}`) and their input is deterministic code over the logged region — reconstructable from log + code, outside the invariant by the unfrozen-request marker.
|
||||
**Model-visible ⟺ logged.** Anything that reaches a model request must be recorded in the session log. The checkable consequence: **every conversation request the loop sends is a pure function of the session log** — anyone holding the log reconstructs it byte-for-byte. Scope, stated precisely: the guarantee covers the loop-built `GenerateOptions`; provider wire bytes follow from it because both adapters' serialization is a pure per-message function at a pinned code version; direct and auxiliary calls own separate log records or deterministic logged inputs — reconstructable from log + code, outside the loop equation because they lack its process-local request identity even when they are frozen.
|
||||
|
||||
Prefix-cache stability is corollary #1, not the headline: an append-only log projected by a per-node pure function yields requests that are append-extensions of their predecessors whenever the header is unchanged — stability is emergent, not managed. Byte-exact audit/replay is corollary #2; resume and fork with *attributable* drift is corollary #3.
|
||||
|
||||
@@ -26,7 +26,7 @@ Each step rebuilds prompt assembly. On the instance's first step, `agent/session
|
||||
|
||||
**`step/start` is the reconstruction boundary.** A step derives messages from events before that sequence. Injection after the snapshot joins the next request, and reentrant appends are rejected during event publication. `agent/pre-step(agent, turn, step, signal)` remains the generic seam for content needed by the current request. Header reconstruction selects the step's `request/header`, or carries the prior snapshot when no new header is written.
|
||||
|
||||
**Enforcement.** In development, `dsh-invariants` independently rebuilds each loop request through a fresh `Session`, so the live cache cannot vouch for itself, then compares messages and folded header fields at `llm/stream`. Loop requests are identified by their frozen shape and session id; direct one-shots are excluded. Correctness depends on sequence-bounded reconstruction rather than listener order. A with-key e2e requires positive cache-read tokens after the first request; per-step usage is the production signal, and a header change or compaction appears as a cache-read drop on the next step.
|
||||
**Enforcement.** In development, `dsh-invariants` independently rebuilds each loop request through a fresh `Session`, so the live cache cannot vouch for itself, then compares messages and folded header fields at `llm/stream`. `dsh-agent-loop` marks the exact request object through `markAgentLoopRequest()`; the marker is process-local and provider-invisible, so immutability plus `sessionId` cannot accidentally classify an auxiliary call as conversation work. Correctness depends on sequence-bounded reconstruction rather than listener order. A with-key e2e requires positive cache-read tokens after the first request; per-step usage is the production signal, and a header change or compaction appears as a cache-read drop on the next step.
|
||||
|
||||
### The MiniCode shape: adopted, with the provenance arrow inverted
|
||||
|
||||
|
||||
@@ -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
|
||||
2026-07-21-log-backed-session-titles.md: 9c54b0b6c24e393f9e7c9f6585d4c80642be9e87
|
||||
2026-07-21-log-backed-session-titles.zh.md: 351acb5d2fd6047547a31ec4a16912ff9ea74a95
|
||||
2026-07-21-log-backed-session-titles.md: 18e7f53c701605971c45c0afff4d7e91d141191e
|
||||
2026-07-21-log-backed-session-titles.zh.md: b9a6b8c7e8c4ca8ae7cb66c1209f9096cc2548cb
|
||||
|
||||
@@ -12,11 +12,11 @@ Session identity metadata is immutable, the event log is the replay and fork bou
|
||||
|
||||
## Decision
|
||||
|
||||
The [`session-title` capability family](../../../../packages/session-title/README.md) owns title state and generation policy. `@deepseek-ai/dsh-session-title` provides `ctx.sessionTitle`, a deterministic first-message fallback, and a registry for at most one optional asynchronous provider. `@deepseek-ai/dsh-session-title-llm` owns the common auxiliary-model request policy; separate first-message and all-user-messages plugins choose input cadence. The shared agent spine mounts only the fallback service with explicit example limits, leaving both model providers opt-in.
|
||||
The [`session-title` capability family](../../../../packages/session-title/README.md) owns title state and generation policy. `@deepseek-ai/dsh-session-title` provides `ctx.sessionTitle`, a deterministic first-message fallback, and a registry for at most one optional asynchronous provider. `@deepseek-ai/dsh-session-title-llm` owns the common auxiliary-model request policy; separate first-message and all-user-messages plugins choose input cadence. The shared agent spine mounts only the fallback service with overridable explicit example limits, leaving both model providers opt-in.
|
||||
|
||||
### Event ownership and folding
|
||||
|
||||
Every accepted revision is a log-only `session/title` event. Its payload contains normalized non-empty text, the exact eligible human `user/message` seqs used to derive it, and either fallback provenance or the registered provider id plus optional provider/model route. Before an auxiliary title-model dispatch, the shared helper appends a log-only `session/title-llm-request` event containing the title-provider id, exact source seqs, route, system prompt, messages, and output-token cap; a later generation failure leaves the request auditable. Validation failures that never reach dispatch create no request event. `foldSessionTitle()` selects the latest title event and adds that event's seq and timestamp as `SessionTitleSnapshot`. Neither event enters `session.surface` or `deriveMessages()`.
|
||||
Every accepted revision is a log-only `session/title` event. Its payload contains normalized non-empty text, the exact eligible human `user/message` seqs used to derive it, and either fallback provenance or the registered provider id plus optional provider/model route. Before an auxiliary title-model dispatch, the shared helper appends a log-only `session/title-llm-request` event containing the title-provider id, exact source seqs, route, system prompt, messages, and output-token cap; a later generation failure leaves the request auditable. The dispatched envelope is deep-frozen to preserve exact agreement with that record but carries no process-local agent-loop request identity, so loop-only reconstruction checks do not compare it with the main conversation header. Validation failures that never reach dispatch create no request event. `foldSessionTitle()` selects the latest title event and adds that event's seq and timestamp as `SessionTitleSnapshot`. Neither event enters `session.surface` or `deriveMessages()`.
|
||||
|
||||
The core session package exposes `ctx.sessions.appendOutOfBand()` only for plugin event types whose owners also declaration-merge an `OutOfBandSessionEventMap` marker. An open turn receives the log-only event directly and owns its normal checkpoint. A closed log receives `turn/start → event → turn/end` under the plugin's trigger, followed by an awaited flush. Once the synthetic turn opens, target-append failure still attempts to close and flush it; detach is deferred until the sequence settles. Session titles contribute the `session-title` zero-step trigger and opt both title event types into this seam.
|
||||
|
||||
@@ -24,7 +24,7 @@ The core session package exposes `ctx.sessions.appendOutOfBand()` only for plugi
|
||||
|
||||
Only text blocks from human-source `user/message` events are eligible. Empty, control-only, and non-text prompts wait for the next eligible message. The service schedules the first fallback without awaiting it from the prompt path, normalizes whitespace and control sequences, applies the configured word and UTF-8 byte limits without splitting a code point, and records the first message seq.
|
||||
|
||||
Automatic provider work starts only after the main loop has a current logged provider/model route. A newly appended `request/header` starts pending work directly; when the header is unchanged, the frozen loop-built `llm/stream` request starts it after matching the folded route. Generation then runs independently of the agent response, and a completion joins whichever turn is open at acceptance time or uses the zero-step append path. Explicit `refresh(session, signal?)` materializes any missing fallback and awaits the registered provider; without a provider it returns the fallback. Caller cancellation during fallback flush does not roll back the durable append, but `refresh()` rechecks the signal and rejects instead of returning success. Concurrent refreshes reserve their session-local revision before waiting for fallback durability, so a newer call supersedes an older call before either can invert provider completion order.
|
||||
Automatic provider work starts only after the main loop has a current logged provider/model route. A newly appended `request/header` starts pending work directly; when the header is unchanged, the marked loop-built `llm/stream` request starts it after matching the folded route. Generation then runs independently of the agent response, and a completion joins whichever turn is open at acceptance time or uses the zero-step append path. Explicit `refresh(session, signal?)` materializes any missing fallback and awaits the registered provider; without a provider it returns the fallback. Caller cancellation during fallback flush does not roll back the durable append, but `refresh()` rechecks the signal and rejects instead of returning success. Concurrent refreshes reserve their session-local revision before waiting for fallback durability, so a newer call supersedes an older call before either can invert provider completion order.
|
||||
|
||||
The first-message provider schedules once when a fresh session first creates its fallback. An automatic failure does not reschedule on later prompts; `refresh()` is the retry path. The all-messages provider schedules after every eligible human prompt and passes all eligible messages through that revision, including seeded history. Its newer revision aborts and supersedes older pending or active work.
|
||||
|
||||
|
||||
@@ -12,11 +12,11 @@ Status: implemented
|
||||
|
||||
## 决策
|
||||
|
||||
[`session-title` 功能包族](../../../../packages/session-title/README.md)负责标题状态和生成策略。`@deepseek-ai/dsh-session-title` 提供 `ctx.sessionTitle`、确定性的首消息回退方案，以及一个至多接受单个可选异步提供方的注册表。`@deepseek-ai/dsh-session-title-llm` 负责通用的辅助模型请求策略；首消息插件和全部用户消息插件分别选择输入调度方式。共享 agent 主干只挂载带显式示例限制的回退服务，两种模型提供方均需按需启用。
|
||||
[`session-title` 功能包族](../../../../packages/session-title/README.md)负责标题状态和生成策略。`@deepseek-ai/dsh-session-title` 提供 `ctx.sessionTitle`、确定性的首消息回退方案，以及一个至多接受单个可选异步提供方的注册表。`@deepseek-ai/dsh-session-title-llm` 负责通用的辅助模型请求策略；首消息插件和全部用户消息插件分别选择输入调度方式。共享 agent 主干只挂载回退服务，并为其显式设置可覆盖的示例限制；两种模型提供方均需按需启用。
|
||||
|
||||
### 事件归属与折叠
|
||||
|
||||
每个已接受的修订都是纯日志 `session/title` 事件。其载荷包含规范化后的非空文本、用于派生标题的所有合格且来源为人类的 `user/message` 的准确 seq，以及回退来源信息，或已注册的提供方 id 加可选的提供方和模型路由。辅助标题模型发起调用前，共享辅助组件会追加一个纯日志 `session/title-llm-request` 事件，其载荷包含标题提供方 id、准确的源 seq、路由、系统提示词、消息和输出 token 上限；即使后续生成失败，这次请求仍可审计。未进入调用阶段的验证失败不会创建请求事件。`foldSessionTitle()` 选择最新的标题事件，并将该事件的 seq 和时间戳加入 `SessionTitleSnapshot`。这两类事件都不会进入 `session.surface` 或 `deriveMessages()`。
|
||||
每个已接受的修订都是纯日志 `session/title` 事件。其载荷包含规范化后的非空文本、用于派生标题的所有合格且来源为人类的 `user/message` 的准确 seq，以及回退来源信息，或已注册的提供方 id 加可选的提供方和模型路由。辅助标题模型发起调用前，共享辅助组件会追加一个纯日志 `session/title-llm-request` 事件，其载荷包含标题提供方 id、准确的源 seq、路由、系统提示词、消息和输出 token 上限；即使后续生成失败，这次请求仍可审计。发送的请求信封经过深度冻结，以确保其与该记录精确一致，但它有意不携带进程本地的 agent loop（智能体循环）请求身份，因此仅针对 agent loop 的重建检查不会将它与主对话请求头进行比较。未进入调用阶段的验证失败不会创建请求事件。`foldSessionTitle()` 选择最新的标题事件，并将该事件的 seq 和时间戳加入 `SessionTitleSnapshot`。这两类事件都不会进入 `session.surface` 或 `deriveMessages()`。
|
||||
|
||||
核心会话包通过 `ctx.sessions.appendOutOfBand()` 暴露这一接口，但只允许所属插件同时通过声明合并向 `OutOfBandSessionEventMap` 添加标记的插件事件类型使用。开放轮次会直接接收纯日志事件，并负责其常规检查点。已关闭的日志会在该插件的触发器下接收 `turn/start → event → turn/end`，随后等待刷写完成。合成轮次一旦开启，即使目标追加失败，系统仍会尝试将其关闭并刷写；整个序列完成前会延迟 detach。会话标题提供 `session-title` 零步骤触发器，并让这两类标题事件都使用这一服务边界。
|
||||
|
||||
@@ -24,7 +24,7 @@ Status: implemented
|
||||
|
||||
只有人类来源的 `user/message` 事件中的文本块才符合条件。空提示词、仅含控制字符的提示词和非文本提示词会等待下一条合格消息。服务从提示词路径调度首个回退标题而不等待其完成，随后规范化空白和控制序列，应用已配置的单词数和 UTF-8 字节限制且不拆分代码点，并记录第一条消息的 seq。
|
||||
|
||||
仅当主循环存在已记录在日志中的当前提供方/模型路由时，自动提供方工作才会启动。`request/header` 新追加到日志时，会直接启动待执行工作；如果请求头没有变化，则由循环构建且已冻结的 `llm/stream` 请求会先与折叠所得的路由匹配，再启动该工作。随后，生成工作独立于 agent 响应运行；完成结果在被接受时加入当时开放的轮次，否则使用零步骤追加路径。显式调用 `refresh(session, signal?)` 会生成尚缺的回退标题并等待已注册的提供方；没有提供方时则返回回退标题。调用方在回退标题刷写期间取消调用不会回滚这次持久化追加，但 `refresh()` 会重新检查取消信号，并让调用以拒绝结束，而非返回成功。并发刷新会在等待回退标题持久化完成前预留会话本地修订号，因此在任何调用有机会造成提供方完成顺序倒置之前，较新的调用就会取代较早的调用。
|
||||
仅当主循环存在已记录在日志中的当前提供方/模型路由时，自动提供方工作才会启动。`request/header` 新追加到日志时，会直接启动待执行工作；如果请求头没有变化，则由循环构建并带有标记的 `llm/stream` 请求会先与折叠所得的路由匹配，再启动该工作。随后，生成工作独立于 agent 响应运行；完成结果在被接受时加入当时开放的轮次，否则使用零步骤追加路径。显式调用 `refresh(session, signal?)` 会生成尚缺的回退标题并等待已注册的提供方；没有提供方时则返回回退标题。调用方在回退标题刷写期间取消调用不会回滚这次持久化追加，但 `refresh()` 会重新检查取消信号，并让调用以拒绝结束，而非返回成功。并发刷新会在等待回退标题持久化完成前预留会话本地修订号，因此在任何调用有机会造成提供方完成顺序倒置之前，较新的调用就会取代较早的调用。
|
||||
|
||||
首消息提供方仅在新会话首次创建回退标题时调度一次。自动执行失败后，后续提示词不会重新调度；`refresh()` 是重试路径。全部消息提供方会在每条合格且由人类发出的提示词后调度，并传入截至该修订的所有合格消息，包括预置历史记录。较新的修订会中止并取代更早的待执行或活跃工作。
|
||||
|
||||
|
||||
@@ -56,6 +56,8 @@ export interface Config {
|
||||
tools?: ToolsConfig
|
||||
/** DeepSeek Harness home directory exposed to bash and used for local skill discovery. */
|
||||
dshHome?: string
|
||||
/** Fallback session-title limits forwarded through agent-spine-demo. */
|
||||
sessionTitle?: NonNullable<agentCore.Config['sessionTitle']>
|
||||
/** Directory the JSONL session backend writes under. Defaults to `./.sessions`. */
|
||||
persistenceRoot?: string
|
||||
/** JSONL artifact encoding; defaults to checksummed Zstandard frames. */
|
||||
@@ -118,7 +120,8 @@ Source: [`packages/core/agent-loop/src/index.ts:360`](../packages/core/agent-loo
|
||||
* bridge, simply omits it), `persona` and `toolOrder` to the system-prompt
|
||||
* plugin (the deployment's persona section and the explicit model-facing tool
|
||||
* order), the `tools` object to the tool registry (its presentation `mode`),
|
||||
* `dshHome` to bash environment and local skill discovery, `skills` to the
|
||||
* `dshHome` to bash environment and local skill discovery, `sessionTitle` to
|
||||
* the fallback title service, `skills` to the
|
||||
* skill registry/local provider/tool consumer, `workspaceContext` to the
|
||||
* workspace-context loader, `llmRetry` to the bounded request-recovery policy,
|
||||
* and `toolBash`/`toolTasks` to the model-facing tool plugins this bundle owns.
|
||||
@@ -142,6 +145,8 @@ export interface Config {
|
||||
tools?: ToolsConfig
|
||||
/** DeepSeek Harness home directory shared by shell context and local skill discovery. */
|
||||
dshHome?: string
|
||||
/** Deterministic fallback and accepted-title limits; omission uses the bundle's example policy. */
|
||||
sessionTitle?: SessionTitleConfig
|
||||
/** Workspace-context loader controls with an explicit byte budget; set `false` for hermetic prompts. */
|
||||
workspaceContext: workspaceContext.Config | false
|
||||
/** Skill registry, local provider, and model-facing consumer config. */
|
||||
@@ -177,9 +182,9 @@ export interface GoalConfig {
|
||||
}
|
||||
```
|
||||
|
||||
Depends on: [`AgentLoopConfig`](#deepseek-aidsh-agent-loop) · [`GoalDomainConfig`](#deepseek-aidsh-goal) · [`llmRetry`](../packages/llm/llm-retry/src/index.ts) · [`SkillLocal`](../packages/skill/skill-local/src/index.ts) · [`SkillRegistryConfig`](#deepseek-aidsh-skill) · [`SystemPromptConfig`](#deepseek-aidsh-system-prompt) · [`toolBash`](../packages/bash/tool-bash/src/index.ts) · [`toolGoal`](../packages/goal/tool-goal/src/index.ts) · [`ToolsConfig`](#deepseek-aidsh-tools) · [`toolSkill`](../packages/skill/tool-skill/src/index.ts) · [`toolTasks`](../packages/tasks/tool-tasks/src/index.ts) · [`workspaceContext`](../packages/context/workspace-context/src/index.ts)
|
||||
Depends on: [`AgentLoopConfig`](#deepseek-aidsh-agent-loop) · [`GoalDomainConfig`](#deepseek-aidsh-goal) · [`llmRetry`](../packages/llm/llm-retry/src/index.ts) · [`SessionTitleConfig`](#deepseek-aidsh-session-title) · [`SkillLocal`](../packages/skill/skill-local/src/index.ts) · [`SkillRegistryConfig`](#deepseek-aidsh-skill) · [`SystemPromptConfig`](#deepseek-aidsh-system-prompt) · [`toolBash`](../packages/bash/tool-bash/src/index.ts) · [`toolGoal`](../packages/goal/tool-goal/src/index.ts) · [`ToolsConfig`](#deepseek-aidsh-tools) · [`toolSkill`](../packages/skill/tool-skill/src/index.ts) · [`toolTasks`](../packages/tasks/tool-tasks/src/index.ts) · [`workspaceContext`](../packages/context/workspace-context/src/index.ts)
|
||||
|
||||
Source: [`packages/examples/agent-spine-demo/src/index.ts:74`](../packages/examples/agent-spine-demo/src/index.ts)
|
||||
Source: [`packages/examples/agent-spine-demo/src/index.ts:82`](../packages/examples/agent-spine-demo/src/index.ts)
|
||||
|
||||
## `@deepseek-ai/dsh-bash-local`
|
||||
|
||||
@@ -242,6 +247,8 @@ export interface Config {
|
||||
tools?: ToolsConfig
|
||||
/** DeepSeek Harness home directory exposed to bash and used for local skill discovery. */
|
||||
dshHome?: string
|
||||
/** Fallback session-title limits forwarded through agent-spine-demo. */
|
||||
sessionTitle?: NonNullable<agentCore.Config['sessionTitle']>
|
||||
/** Directory the JSONL session backend writes under. Defaults to `./.sessions`. */
|
||||
persistenceRoot?: string
|
||||
/** JSONL artifact encoding; defaults to checksummed Zstandard frames. */
|
||||
@@ -1404,6 +1411,8 @@ export interface Config {
|
||||
tools?: ToolsConfig
|
||||
/** DeepSeek Harness home directory exposed to bash and used for local skill discovery. */
|
||||
dshHome?: string
|
||||
/** Fallback session-title limits forwarded through agent-spine-demo. */
|
||||
sessionTitle?: NonNullable<agentCore.Config['sessionTitle']>
|
||||
/** Directory the JSONL session backend writes under. Defaults to `./.sessions`. */
|
||||
persistenceRoot?: string
|
||||
/** JSONL artifact encoding; defaults to checksummed Zstandard frames. */
|
||||
|
||||
@@ -524,11 +524,11 @@ Waterfall around every streaming model call (retry, replay, routing). Bound to t
|
||||
* Waterfall around every streaming model call (retry, replay, routing).
|
||||
* Bound to the {@link LlmService}; call `next()` to reach the resolved
|
||||
* adapter's stream, or yield your own chunks to short-circuit.
|
||||
* @param options - the full request. A LOOP-built request arrives
|
||||
* deep-frozen (mutation throws): its content is a pure function of the
|
||||
* session log (the reconstructability Agent Note), so listeners read it, never
|
||||
* rewrite it. A hand-built one-shot (compaction summarize) is the
|
||||
* caller's own object and stays mutable here.
|
||||
* @param options - the full request. A LOOP-built request carries the
|
||||
* process-local {@link markAgentLoopRequest} identity and arrives deep-frozen
|
||||
* (mutation throws): its content is a pure function of the session log (the
|
||||
* reconstructability Agent Note), so listeners read it, never rewrite it.
|
||||
* Hand-built calls own their mutability policy and do not carry that marker.
|
||||
* @mode waterfall
|
||||
*/
|
||||
'llm/stream'(this: LlmService, options: GenerateOptions, next: () => AsyncIterable<StreamChunk>): AsyncIterable<StreamChunk>
|
||||
|
||||
@@ -269,7 +269,7 @@ The model-facing `ToolSchema` is the wire shape; the registered `ToolDefinition`
|
||||
|
||||
The loop builds each request from logged state. `EpochHeader` records call config, rendered prompt, authoritative returned tool order (configured by `toolOrder`, or lexicographic when unset), and session prefix through full `request/header` snapshots. Together with derived history, this makes the request reconstructable from the session log. See [session.md](session.md#the-request-header-event-requestheader) and the [reconstructability Agent Note](../../.agents/notes/implemented/architecture/2026-07-05-reconstructable-requests.md).
|
||||
|
||||
`agent/request` receives a frozen call-config seed and may return a replacement to switch provider, model, or sampling. `agent/session-prefix` composes request-only prefix messages once per loop instance, and the header records the exact result used. Requests reaching `llm/stream` are deep-frozen, so mutation throws.
|
||||
`agent/request` receives a frozen call-config seed and may return a replacement to switch provider, model, or sampling. `agent/session-prefix` composes request-only prefix messages once per loop instance, and the header records the exact result used. Requests reaching `llm/stream` are deep-frozen, so mutation throws, and carry a process-local loop identity so observers do not confuse separately logged frozen auxiliary calls with conversation requests.
|
||||
|
||||
On the wire, a loop-built request reads in this order: the `system` slot (the rendered prompt assembly) → `messagePrefix` (the frozen session prefix) → the derived history — the boundary snapshot, whose tail is the newest `user/message` on a turn's first step and the previous step's tool results on later steps. The prefix never enters the derived history; its durable record is the header events, and the dev invariant recomputes exactly this equation against every loop-built request.
|
||||
|
||||
|
||||
@@ -884,7 +884,7 @@ export const EVENT_API: readonly EventApiEntry[] = [
|
||||
name: 'llm/stream',
|
||||
mode: 'waterfall',
|
||||
signature: '\'llm/stream\'(this: LlmService, options: GenerateOptions, next: () => AsyncIterable<StreamChunk>): AsyncIterable<StreamChunk>',
|
||||
jsDoc: '/**\n * Waterfall around every streaming model call (retry, replay, routing).\n * Bound to the {@link LlmService}; call `next()` to reach the resolved\n * adapter\'s stream, or yield your own chunks to short-circuit.\n * @param options - the full request. A LOOP-built request arrives\n * deep-frozen (mutation throws): its content is a pure function of the\n * session log (the reconstructability Agent Note), so listeners read it, never\n * rewrite it. A hand-built one-shot (compaction summarize) is the\n * caller\'s own object and stays mutable here.\n * @mode waterfall\n */',
|
||||
jsDoc: '/**\n * Waterfall around every streaming model call (retry, replay, routing).\n * Bound to the {@link LlmService}; call `next()` to reach the resolved\n * adapter\'s stream, or yield your own chunks to short-circuit.\n * @param options - the full request. A LOOP-built request carries the\n * process-local {@link markAgentLoopRequest} identity and arrives deep-frozen\n * (mutation throws): its content is a pure function of the session log (the\n * reconstructability Agent Note), so listeners read it, never rewrite it.\n * Hand-built calls own their mutability policy and do not carry that marker.\n * @mode waterfall\n */',
|
||||
summary: 'Waterfall around every streaming model call (retry, replay, routing).',
|
||||
},
|
||||
{
|
||||
|
||||
@@ -8,7 +8,7 @@
|
||||
import type { Context } from 'cordis'
|
||||
import type { ContentBlock, FinishReason, GenerateOptions, LlmCallConfig, LlmFailure, Message } from '@deepseek-ai/dsh-llm'
|
||||
import { isDeepStrictEqual } from 'node:util'
|
||||
import { BlockAssembler, HarnessError, LlmError, assertNever, deepFreeze, errorChain, llmFailureOf } from '@deepseek-ai/dsh-llm'
|
||||
import { BlockAssembler, HarnessError, LlmError, assertNever, deepFreeze, errorChain, llmFailureOf, markAgentLoopRequest } from '@deepseek-ai/dsh-llm'
|
||||
import { agentEvents, assembleContextFor } from '@deepseek-ai/dsh-agent'
|
||||
import type { AgentEventDispatch, ContinuationDecision, HookContext, PromptDecision, RequestError, RequestErrorDecision } from '@deepseek-ai/dsh-agent'
|
||||
import { canonicalHeader } from '@deepseek-ai/dsh-session'
|
||||
@@ -619,7 +619,7 @@ async function runStep(
|
||||
recordRequestHeader(session, transmission, header)
|
||||
|
||||
// Freeze the logged header plus boundary snapshot; the prefix precedes derived history.
|
||||
const request: GenerateOptions = deepFreeze({
|
||||
const request: GenerateOptions = markAgentLoopRequest(deepFreeze({
|
||||
provider: header.config.provider,
|
||||
model: header.config.model,
|
||||
messages: [...header.messagePrefix ?? [], ...boundaryMessages],
|
||||
@@ -630,7 +630,7 @@ async function runStep(
|
||||
...header.config.stop !== undefined ? { stop: header.config.stop } : {},
|
||||
sessionId: session.id,
|
||||
signal,
|
||||
})
|
||||
}))
|
||||
|
||||
// --- Model call (streaming-first; raw chunks are the replay record) ---
|
||||
const assembler = new BlockAssembler()
|
||||
|
||||
@@ -33,6 +33,7 @@ Because the package wires no logger entry, an ACP leaf has **nothing to get wron
|
||||
| `persona` | — | the deployment persona template (may reference `{{provider}}`/`{{model}}`/`{{cwd}}`), routed to `dsh-system-prompt` |
|
||||
| `toolOrder` | — | explicit model-facing tool order (a name list with one `'<unlisted-tools>'` rest entry; absent — lexicographic; an unregistered name fails each turn at prompt assembly), routed to `dsh-system-prompt` |
|
||||
| `dshHome` | `$DSH_HOME` or `~/.dsh` | Harness home exposed to model bash and used by local skill discovery |
|
||||
| `sessionTitle` | spine example limits | fallback title word/byte limits routed through `dsh-agent-spine-demo` |
|
||||
| `tools` | `{ mode: 'native' }` | tool-registry presentation config (`native` / `code` / `both`), routed through `dsh-agent-spine-demo` |
|
||||
| `workspaceContext` | (required) | workspace-instruction byte budget/config, or `false`; routed to the providerless-safe `dsh-workspace-context` plugin |
|
||||
| `skills` | owner defaults | registry-cache, local-provider, and model-facing skill-tool config, routed through `dsh-agent-spine-demo` |
|
||||
|
||||
@@ -50,6 +50,8 @@ export interface Config {
|
||||
tools?: ToolsConfig
|
||||
/** DeepSeek Harness home directory exposed to bash and used for local skill discovery. */
|
||||
dshHome?: string
|
||||
/** Fallback session-title limits forwarded through agent-spine-demo. */
|
||||
sessionTitle?: NonNullable<agentCore.Config['sessionTitle']>
|
||||
/** Directory the JSONL session backend writes under. Defaults to `./.sessions`. */
|
||||
persistenceRoot?: string
|
||||
/** JSONL artifact encoding; defaults to checksummed Zstandard frames. */
|
||||
@@ -82,6 +84,7 @@ export const Config: z<Config> = z.object({
|
||||
toolOrder: z.array(z.string()).default(undefined as unknown as string[]),
|
||||
tools: ToolRegistry.Config,
|
||||
dshHome: z.string(),
|
||||
sessionTitle: agentCore.SessionTitleConfigSchema,
|
||||
persistenceRoot: z.string().default(DEFAULT_PERSISTENCE_ROOT),
|
||||
persistenceCompression: JsonlCompressionSchema,
|
||||
workspaceContext: z.union([z.const(false), workspaceContext.Config]).required(),
|
||||
|
||||
@@ -37,7 +37,7 @@ Read this package for the whole plugin tree and its composition order.
|
||||
The spine is everything COMMON to every front door. The swappable and front-door-coupled pieces stay out, picked by whatever loads the bundle:
|
||||
|
||||
- **the LLM adapter** — the bundle ships the abstract `llm` service; the leaf registers a concrete adapter on `ctx.llm` (`llm-deepseek`, `llm-pi-ai`, `llm-replay`).
|
||||
- **model-backed session-title providers** — the bundle mounts the fallback service with fixed example limits (5 words, 40 fallback bytes, 80 accepted-title bytes); a leaf may opt into exactly one first-message or all-messages LLM provider.
|
||||
- **model-backed session-title providers** — the bundle mounts the fallback service with overridable example limits (5 words, 40 fallback bytes, 80 accepted-title bytes); a leaf may opt into exactly one first-message or all-messages LLM provider.
|
||||
- **the bash executor** — the bundle ships `tool-bash` (the consumer schema); the leaf provides `ctx.bash` (`bash-local` or a sandboxed impl).
|
||||
- **non-local skill providers** — the bundle ships the skill registry, the local filesystem provider, and the `skill` tool; deployments can add other providers such as embedded or remote catalogs as siblings.
|
||||
- **presentation + per-app infra** — the terminal TUI or ACP front door and `hmr`. These form the coupled front-door cluster that the app packages ([`dsh-tui-demo`](../tui-demo/README.md), [`dsh-acp-demo`](../acp-demo/README.md)) bake in. `timer` is in the spine because it is common and stdout-silent; front doors own stdout and remain outside.
|
||||
@@ -48,11 +48,11 @@ This is the [interface/implementation/consumer seam](../../../.agents/notes/impl
|
||||
|
||||
```ts
|
||||
import type { Config } from '@deepseek-ai/dsh-agent-spine-demo'
|
||||
// { agents?, maxParallelToolCalls?, persona?, toolOrder?, tools?, dshHome?, skills?, workspaceContext, toolBash?, toolTasks?, goals?, llmRetry? }
|
||||
// { agents?, maxParallelToolCalls?, persona?, toolOrder?, tools?, dshHome?, sessionTitle?, skills?, workspaceContext, toolBash?, toolTasks?, goals?, llmRetry? }
|
||||
// workspaceContext requires { maxBytes } or false; the other owner schemas supply defaults.
|
||||
```
|
||||
|
||||
The bundle FORWARDS each field to the child that owns it: `agents` and `maxParallelToolCalls` to `agent-loop` (`agents` defaults to `[]`; the cap defaults there), so each app supplies its own pre-created agents — TUI and headless apps pre-create `main`, while the ACP app creates agents on demand at `session/new`; `llmRetry` to the bounded retry policy; `persona` and `toolOrder` to `dsh-system-prompt`; `tools` to the tool registry for its presentation mode; `skills.registry`, `skills.local`, and `skills.tool` to the skill registry, local provider, and model-facing consumer; the required `workspaceContext` choice to `dsh-workspace-context` (`{ maxBytes }` enables loading and `false` disables it); and `toolBash`/`toolTasks` to the two model-facing tool plugins the bundle owns. The session-title fallback limits are fixed example composition policy rather than bundle config. A `goals` object opts into the persisted domain, model tools, and same-session driver while forwarding `goals.domain` and `goals.tool` to their owners; omission or `false` leaves the stack absent so headless callers retain one-turn settlement. Set `skills.enabled: false` to omit both the local provider and model-facing skill tool, and set `toolTasks: false` to retain the task service for foreground producers without exposing `task_output` / `task_list` / `task_kill`. It resolves `dshHome` once through [`@deepseek-ai/dsh-home`](../../util/home/README.md) and forwards that absolute value to tool-bash's managed environment and enabled local skill discovery. An absent top-level `dshHome` adopts `skills.local.dshHome`; supplying both with different resolved paths fails loudly. `toolBash.enableRunInBackground` controls only the bash producer; independently loaded producers keep their own config. Workspace instructions register before the skill catalog so their session-prefix message renders first. App packages use `pickSpineConfig()` to copy only these bundle-owned fields.
|
||||
The bundle FORWARDS each field to the child that owns it: `agents` and `maxParallelToolCalls` to `agent-loop` (`agents` defaults to `[]`; the cap defaults there), so each app supplies its own pre-created agents — TUI and headless apps pre-create `main`, while the ACP app creates agents on demand at `session/new`; `llmRetry` to the bounded retry policy; `persona` and `toolOrder` to `dsh-system-prompt`; `tools` to the tool registry for its presentation mode; `sessionTitle` to the fallback title service; `skills.registry`, `skills.local`, and `skills.tool` to the skill registry, local provider, and model-facing consumer; the required `workspaceContext` choice to `dsh-workspace-context` (`{ maxBytes }` enables loading and `false` disables it); and `toolBash`/`toolTasks` to the two model-facing tool plugins the bundle owns. Omitted `sessionTitle` uses the explicit example policy of 5 words, 40 fallback bytes, and 80 accepted-title bytes. A `goals` object opts into the persisted domain, model tools, and same-session driver while forwarding `goals.domain` and `goals.tool` to their owners; omission or `false` leaves the stack absent so headless callers retain one-turn settlement. Set `skills.enabled: false` to omit both the local provider and model-facing skill tool, and set `toolTasks: false` to retain the task service for foreground producers without exposing `task_output` / `task_list` / `task_kill`. It resolves `dshHome` once through [`@deepseek-ai/dsh-home`](../../util/home/README.md) and forwards that absolute value to tool-bash's managed environment and enabled local skill discovery. An absent top-level `dshHome` adopts `skills.local.dshHome`; supplying both with different resolved paths fails loudly. `toolBash.enableRunInBackground` controls only the bash producer; independently loaded producers keep their own config. Workspace instructions register before the skill catalog so their session-prefix message renders first. App packages use `pickSpineConfig()` to copy only these bundle-owned fields.
|
||||
|
||||
## Why a code bundle, not a shared YAML include
|
||||
|
||||
|
||||
@@ -13,7 +13,7 @@ import Timer from '@cordisjs/plugin-timer'
|
||||
import z from 'schemastery'
|
||||
import LlmService from '@deepseek-ai/dsh-llm'
|
||||
import SessionStore from '@deepseek-ai/dsh-session'
|
||||
import SessionTitleService from '@deepseek-ai/dsh-session-title'
|
||||
import SessionTitleService, { type Config as SessionTitleConfig } from '@deepseek-ai/dsh-session-title'
|
||||
import SystemPrompt, { type Config as SystemPromptConfig } from '@deepseek-ai/dsh-system-prompt'
|
||||
import ToolRegistry, { type Config as ToolsConfig } from '@deepseek-ai/dsh-tools'
|
||||
import SkillService, { type Config as SkillRegistryConfig } from '@deepseek-ai/dsh-skill'
|
||||
@@ -34,6 +34,13 @@ import { resolveDshHome } from '@deepseek-ai/dsh-home'
|
||||
|
||||
export const name = 'agent-spine-demo'
|
||||
|
||||
/** Overridable example policy used when a bundle consumer omits `sessionTitle`. */
|
||||
const EXAMPLE_SESSION_TITLE_CONFIG: SessionTitleConfig = {
|
||||
fallbackMaxWords: 5,
|
||||
fallbackMaxBytes: 40,
|
||||
maxTitleBytes: 80,
|
||||
}
|
||||
|
||||
/** Skill bundle config forwarded to the registry, local provider, and model-facing consumer. */
|
||||
export interface SkillConfig {
|
||||
/** Mount the bundled local skill provider and model-facing skill tool (default true). */
|
||||
@@ -60,7 +67,8 @@ export interface GoalConfig {
|
||||
* bridge, simply omits it), `persona` and `toolOrder` to the system-prompt
|
||||
* plugin (the deployment's persona section and the explicit model-facing tool
|
||||
* order), the `tools` object to the tool registry (its presentation `mode`),
|
||||
* `dshHome` to bash environment and local skill discovery, `skills` to the
|
||||
* `dshHome` to bash environment and local skill discovery, `sessionTitle` to
|
||||
* the fallback title service, `skills` to the
|
||||
* skill registry/local provider/tool consumer, `workspaceContext` to the
|
||||
* workspace-context loader, `llmRetry` to the bounded request-recovery policy,
|
||||
* and `toolBash`/`toolTasks` to the model-facing tool plugins this bundle owns.
|
||||
@@ -84,6 +92,8 @@ export interface Config {
|
||||
tools?: ToolsConfig
|
||||
/** DeepSeek Harness home directory shared by shell context and local skill discovery. */
|
||||
dshHome?: string
|
||||
/** Deterministic fallback and accepted-title limits; omission uses the bundle's example policy. */
|
||||
sessionTitle?: SessionTitleConfig
|
||||
/** Workspace-context loader controls with an explicit byte budget; set `false` for hermetic prompts. */
|
||||
workspaceContext: workspaceContext.Config | false
|
||||
/** Skill registry, local provider, and model-facing consumer config. */
|
||||
@@ -106,6 +116,10 @@ export const SkillConfigSchema: z<SkillConfig> = z.object({
|
||||
tool: toolSkill.Config,
|
||||
})
|
||||
|
||||
/** The session-title config schema with the shared bundle's overridable example limits. */
|
||||
export const SessionTitleConfigSchema: z<SessionTitleConfig> = SessionTitleService.Config
|
||||
.default(EXAMPLE_SESSION_TITLE_CONFIG)
|
||||
|
||||
/** The bash-tool config schema exported for app packages that forward `toolBash`. */
|
||||
export const ToolBashConfigSchema: z<toolBash.Config> = toolBash.Config
|
||||
|
||||
@@ -128,13 +142,14 @@ export const Config = z.intersect([
|
||||
z.object({
|
||||
tools: ToolRegistry.Config,
|
||||
dshHome: z.string(),
|
||||
sessionTitle: SessionTitleConfigSchema,
|
||||
skills: SkillConfigSchema,
|
||||
workspaceContext: z.union([z.const(false), workspaceContext.Config]).required(),
|
||||
toolBash: ToolBashConfigSchema,
|
||||
toolTasks: z.union([z.const(false), ToolTasksConfigSchema]),
|
||||
goals: z.union([z.const(false), GoalConfigSchema]),
|
||||
llmRetry: LlmRetryConfigSchema,
|
||||
}) as unknown as z<Pick<Config, 'tools' | 'dshHome' | 'skills' | 'workspaceContext' | 'toolBash' | 'toolTasks' | 'goals' | 'llmRetry'>>,
|
||||
}) as unknown as z<Pick<Config, 'tools' | 'dshHome' | 'sessionTitle' | 'skills' | 'workspaceContext' | 'toolBash' | 'toolTasks' | 'goals' | 'llmRetry'>>,
|
||||
]) as unknown as z<Config>
|
||||
|
||||
/**
|
||||
@@ -149,6 +164,7 @@ export function pickSpineConfig(config: Omit<Config, 'agents'>): Omit<Config, 'a
|
||||
...config.toolOrder !== undefined ? { toolOrder: config.toolOrder } : {},
|
||||
...config.tools !== undefined ? { tools: config.tools } : {},
|
||||
...config.dshHome !== undefined ? { dshHome: config.dshHome } : {},
|
||||
...config.sessionTitle !== undefined ? { sessionTitle: config.sessionTitle } : {},
|
||||
workspaceContext: config.workspaceContext,
|
||||
...config.skills !== undefined ? { skills: config.skills } : {},
|
||||
...config.toolBash !== undefined ? { toolBash: config.toolBash } : {},
|
||||
@@ -179,11 +195,7 @@ export function apply(ctx: Context, config: Config): void {
|
||||
ctx.plugin(Timer)
|
||||
ctx.plugin(LlmService)
|
||||
ctx.plugin(SessionStore)
|
||||
ctx.plugin(SessionTitleService, {
|
||||
fallbackMaxWords: 5,
|
||||
fallbackMaxBytes: 40,
|
||||
maxTitleBytes: 80,
|
||||
})
|
||||
ctx.plugin(SessionTitleService, config.sessionTitle ?? EXAMPLE_SESSION_TITLE_CONFIG)
|
||||
// Owner schemas resolve defaults; forward toolOrder only when explicitly set.
|
||||
ctx.plugin(SystemPrompt, {
|
||||
persona: config.persona ?? '',
|
||||
|
||||
@@ -129,6 +129,30 @@ describe('dsh-agent-spine-demo bundle', () => {
|
||||
await ctx.fiber.dispose()
|
||||
})
|
||||
|
||||
it('forwards configurable fallback title limits to the bundled service', async () => {
|
||||
const ctx = await mount({
|
||||
workspaceContext: false,
|
||||
sessionTitle: {
|
||||
fallbackMaxWords: 1,
|
||||
fallbackMaxBytes: 40,
|
||||
maxTitleBytes: 80,
|
||||
},
|
||||
})
|
||||
const session = ctx.sessions.create(SessionId('configured-title-limits'))
|
||||
session.append('turn/start', {
|
||||
turn: 1,
|
||||
trigger: { kind: 'message', source: { kind: 'user' } },
|
||||
})
|
||||
session.append('user/message', {
|
||||
content: [{ type: 'text', text: 'One two three four' }],
|
||||
source: { kind: 'user' },
|
||||
}, { surfaceOp: 'append' })
|
||||
await new Promise(resolve => setTimeout(resolve, 0))
|
||||
|
||||
expect(ctx.sessionTitle.get(session)?.title).toBe('One')
|
||||
await ctx.fiber.dispose()
|
||||
})
|
||||
|
||||
it('opts into the configured persisted-goal domain, tools, and same-session driver', async () => {
|
||||
const ctx = await mount({
|
||||
workspaceContext: false,
|
||||
@@ -455,6 +479,7 @@ describe('dsh-agent-spine-demo bundle', () => {
|
||||
toolOrder: ['zulu'],
|
||||
tools: { mode: 'native' as const },
|
||||
dshHome: '/tmp/dsh-home',
|
||||
sessionTitle: { fallbackMaxWords: 3, fallbackMaxBytes: 24, maxTitleBytes: 60 },
|
||||
workspaceContext: false as const,
|
||||
skills: { enabled: false },
|
||||
toolBash: { enableRunInBackground: false },
|
||||
@@ -467,6 +492,7 @@ describe('dsh-agent-spine-demo bundle', () => {
|
||||
toolOrder: appConfig.toolOrder,
|
||||
tools: appConfig.tools,
|
||||
dshHome: appConfig.dshHome,
|
||||
sessionTitle: appConfig.sessionTitle,
|
||||
workspaceContext: false,
|
||||
skills: appConfig.skills,
|
||||
toolBash: appConfig.toolBash,
|
||||
|
||||
@@ -15,6 +15,7 @@ The package mounts no console logger, interactive UI, user-interaction service,
|
||||
| `toolOrder` | lexicographic | explicit model-facing tool order in `dsh-system-prompt` |
|
||||
| `tools` | `{ mode: 'native' }` | tool-registry presentation config through `dsh-agent-spine-demo` |
|
||||
| `dshHome` | `$DSH_HOME` or `~/.dsh` | Harness home exposed to model bash and used by local skill discovery |
|
||||
| `sessionTitle` | spine example limits | Fallback title word/byte limits through `dsh-agent-spine-demo` |
|
||||
| `skills` | owner defaults | skill registry, local provider, and model-facing skill tool |
|
||||
| `toolBash` | owner defaults | model-facing bash config, including this producer's background opt-in |
|
||||
| `toolTasks` | owner defaults | generic `task_output` wait bounds |
|
||||
|
||||
@@ -37,6 +37,8 @@ export interface Config {
|
||||
tools?: ToolsConfig
|
||||
/** DeepSeek Harness home directory exposed to bash and used for local skill discovery. */
|
||||
dshHome?: string
|
||||
/** Fallback session-title limits forwarded through agent-spine-demo. */
|
||||
sessionTitle?: NonNullable<agentCore.Config['sessionTitle']>
|
||||
/** Directory the JSONL session backend writes under. Defaults to `./.sessions`. */
|
||||
persistenceRoot?: string
|
||||
/** JSONL artifact encoding; defaults to checksummed Zstandard frames. */
|
||||
@@ -64,6 +66,7 @@ export const Config: z<Config> = z.object({
|
||||
persistenceCompression: JsonlCompressionSchema,
|
||||
persona: z.string(),
|
||||
dshHome: z.string(),
|
||||
sessionTitle: agentCore.SessionTitleConfigSchema,
|
||||
skills: agentCore.SkillConfigSchema,
|
||||
// Absent means lexicographic order; schemastery's native array default is [].
|
||||
toolOrder: z.array(z.string()).default(undefined as unknown as string[]),
|
||||
|
||||
@@ -29,6 +29,7 @@ Swappable LLM, bash, filesystem, and other capability providers remain in the le
|
||||
| `toolOrder` | lexicographic | Explicit model-facing tool order |
|
||||
| `tools` | owner default | Tool presentation mode |
|
||||
| `dshHome` | owner default | Harness home used by bash and skills |
|
||||
| `sessionTitle` | spine example limits | Fallback title word/byte limits |
|
||||
| `skills` | owner defaults | Skill registry, local provider, and tool config |
|
||||
| `toolBash` | owner defaults | Model-facing bash tool config |
|
||||
| `toolTasks` | owner defaults | Background-task control-tool config, or `false` |
|
||||
|
||||
@@ -45,6 +45,8 @@ export interface Config {
|
||||
tools?: ToolsConfig
|
||||
/** DeepSeek Harness home directory exposed to bash and used for local skill discovery. */
|
||||
dshHome?: string
|
||||
/** Fallback session-title limits forwarded through agent-spine-demo. */
|
||||
sessionTitle?: NonNullable<agentCore.Config['sessionTitle']>
|
||||
/** Directory the JSONL session backend writes under. Defaults to `./.sessions`. */
|
||||
persistenceRoot?: string
|
||||
/** JSONL artifact encoding; defaults to checksummed Zstandard frames. */
|
||||
@@ -79,6 +81,7 @@ export const Config: z<Config> = z.object({
|
||||
toolOrder: z.array(z.string()).default(undefined as unknown as string[]),
|
||||
tools: ToolRegistry.Config,
|
||||
dshHome: z.string(),
|
||||
sessionTitle: agentCore.SessionTitleConfigSchema,
|
||||
persistenceRoot: z.string().default(DEFAULT_PERSISTENCE_ROOT),
|
||||
persistenceCompression: JsonlCompressionSchema,
|
||||
welcome: z.string().default(DEFAULT_WELCOME),
|
||||
|
||||
@@ -36,7 +36,7 @@ Streaming is a raw chunk protocol (`block-start`, `text-delta`, `reasoning-delta
|
||||
|
||||
### Call configuration (`call-config.ts`)
|
||||
|
||||
`LlmCallConfig` is the provider + model + sampling scalars of one conversation's requests (`provider`, `model`, `temperature`, `maxTokens`, `stop` — each mapping 1:1 onto the same-named `GenerateOptions` field). It is per-conversation state recorded in the session log as part of the request header (see the dsh-session `request/header` events), never a silently-adjustable per-call knob: the `agent/request` waterfall proposes a replacement and the loop logs a real change. `callConfigEquals(a, b)` is the field-wise real-change detector; `deepFreeze(value)` is the ownership helper the loop applies to every built request before dispatch (`llm/stream` listeners and adapters read, never rewrite).
|
||||
`LlmCallConfig` is the provider + model + sampling scalars of one conversation's requests (`provider`, `model`, `temperature`, `maxTokens`, `stop` — each mapping 1:1 onto the same-named `GenerateOptions` field). It is per-conversation state recorded in the session log as part of the request header (see the dsh-session `request/header` events), never a silently-adjustable per-call knob: the `agent/request` waterfall proposes a replacement and the loop logs a real change. `callConfigEquals(a, b)` is the field-wise real-change detector; `deepFreeze(value)` is the ownership helper the loop applies to every built request before dispatch (`llm/stream` listeners and adapters read, never rewrite). `markAgentLoopRequest()` gives that exact object process-local loop provenance, and `isAgentLoopRequest()` lets observers distinguish it from independently logged auxiliary calls that may also be frozen and session-associated.
|
||||
|
||||
### App attribution (`attribution.ts`)
|
||||
|
||||
|
||||
@@ -6,6 +6,11 @@
|
||||
* @module dsh-llm/call-config
|
||||
*/
|
||||
|
||||
import type { GenerateOptions } from './types.ts'
|
||||
|
||||
/** Process-local identities of request objects assembled by dsh-agent-loop. */
|
||||
const AGENT_LOOP_REQUESTS = new WeakSet<GenerateOptions>()
|
||||
|
||||
/**
|
||||
* Provider + model + sampling scalars of one conversation's requests. Every field maps
|
||||
* 1:1 onto the same-named `GenerateOptions` field; the loop builds requests
|
||||
@@ -33,6 +38,25 @@ export function callConfigEquals(a: LlmCallConfig, b: LlmCallConfig): boolean {
|
||||
return a.stop.length === b.stop.length && a.stop.every((s, i) => s === b.stop?.[i])
|
||||
}
|
||||
|
||||
/**
|
||||
* Mark one exact request object as assembled by dsh-agent-loop.
|
||||
* @param request - loop-owned request envelope before LLM dispatch.
|
||||
* @returns the same request object with process-local loop provenance.
|
||||
*/
|
||||
export function markAgentLoopRequest<T extends GenerateOptions>(request: T): T {
|
||||
AGENT_LOOP_REQUESTS.add(request)
|
||||
return request
|
||||
}
|
||||
|
||||
/**
|
||||
* Test whether the exact request object was assembled by dsh-agent-loop.
|
||||
* @param request - request envelope observed at the LLM waterfall.
|
||||
* @returns whether {@link markAgentLoopRequest} recorded this object.
|
||||
*/
|
||||
export function isAgentLoopRequest(request: GenerateOptions): boolean {
|
||||
return AGENT_LOOP_REQUESTS.has(request)
|
||||
}
|
||||
|
||||
/**
|
||||
* Deep-freeze a value in place, guarding cycles, so later mutation throws.
|
||||
* {@link AbortSignal} objects are deliberately skipped because they are the
|
||||
|
||||
@@ -20,7 +20,7 @@ export * from './never.ts'
|
||||
export * from './error.ts'
|
||||
export * from './types.ts'
|
||||
export { BlockAssembler } from './assembler.ts'
|
||||
export { callConfigEquals, deepFreeze } from './call-config.ts'
|
||||
export { callConfigEquals, deepFreeze, isAgentLoopRequest, markAgentLoopRequest } from './call-config.ts'
|
||||
export type { LlmCallConfig } from './call-config.ts'
|
||||
export { isLlmAdapterFailure, llmFailureOf } from './adapter-failure.ts'
|
||||
|
||||
@@ -34,11 +34,11 @@ declare module 'cordis' {
|
||||
* Waterfall around every streaming model call (retry, replay, routing).
|
||||
* Bound to the {@link LlmService}; call `next()` to reach the resolved
|
||||
* adapter's stream, or yield your own chunks to short-circuit.
|
||||
* @param options - the full request. A LOOP-built request arrives
|
||||
* deep-frozen (mutation throws): its content is a pure function of the
|
||||
* session log (the reconstructability Agent Note), so listeners read it, never
|
||||
* rewrite it. A hand-built one-shot (compaction summarize) is the
|
||||
* caller's own object and stays mutable here.
|
||||
* @param options - the full request. A LOOP-built request carries the
|
||||
* process-local {@link markAgentLoopRequest} identity and arrives deep-frozen
|
||||
* (mutation throws): its content is a pure function of the session log (the
|
||||
* reconstructability Agent Note), so listeners read it, never rewrite it.
|
||||
* Hand-built calls own their mutability policy and do not carry that marker.
|
||||
* @mode waterfall
|
||||
*/
|
||||
'llm/stream'(this: LlmService, options: GenerateOptions, next: () => AsyncIterable<StreamChunk>): AsyncIterable<StreamChunk>
|
||||
|
||||
@@ -5,7 +5,8 @@
|
||||
*/
|
||||
|
||||
import { describe, expect, it } from 'vitest'
|
||||
import { callConfigEquals, deepFreeze } from '../src/call-config.ts'
|
||||
import { callConfigEquals, deepFreeze, isAgentLoopRequest, markAgentLoopRequest } from '../src/call-config.ts'
|
||||
import type { GenerateOptions } from '../src/types.ts'
|
||||
|
||||
describe('callConfigEquals', () => {
|
||||
it('compares every field, including the stop list element-wise', () => {
|
||||
@@ -56,3 +57,19 @@ describe('deepFreeze', () => {
|
||||
expect(Object.isFrozen(cyclic)).toBe(true)
|
||||
})
|
||||
})
|
||||
|
||||
describe('agent-loop request identity', () => {
|
||||
it('marks only the exact request object and preserves its identity', () => {
|
||||
const request: GenerateOptions = {
|
||||
provider: 'mock',
|
||||
model: 'model',
|
||||
messages: [],
|
||||
}
|
||||
const copy = { ...request }
|
||||
|
||||
expect(isAgentLoopRequest(request)).toBe(false)
|
||||
expect(markAgentLoopRequest(request)).toBe(request)
|
||||
expect(isAgentLoopRequest(request)).toBe(true)
|
||||
expect(isAgentLoopRequest(copy)).toBe(false)
|
||||
})
|
||||
})
|
||||
|
||||
@@ -8,7 +8,7 @@ This package is a library, not a Cordis plugin. The provider plugins call `regis
|
||||
|
||||
`provider` and `model` overrides are optional but must be supplied together as non-empty strings. Without that pair, the helper uses the exact provider/model route captured from the current session's logged `request/header`; an explicit refresh before any route exists therefore needs overrides. Input exceeding `maxInputBytes` rejects instead of being truncated. Timeout, cancellation, malformed or empty output, tool calls, and non-stop finish reasons also reject; the session-title service decides whether that rejection is an automatic warning or an explicit caller failure.
|
||||
|
||||
After route and input validation, the helper appends a log-only `session/title-llm-request` event before model dispatch. It contains the title-provider id, exact source seqs, route, system prompt, message list, and output-token cap used by the call. A later model failure leaves that request record intact; validation failures that never become dispatchable requests do not create one. The event stays outside derived model history.
|
||||
After route and input validation, the helper appends a log-only `session/title-llm-request` event before model dispatch. It contains the title-provider id, exact source seqs, route, system prompt, message list, and output-token cap used by the call. The dispatched envelope is deep-frozen to keep interceptors aligned with that record but deliberately lacks dsh-agent-loop's process-local request identity, so loop-only reconstruction observers do not compare it with the conversation header. A later model failure leaves that request record intact; validation failures that never become dispatchable requests do not create one. The event stays outside derived model history.
|
||||
|
||||
## Configuration
|
||||
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
import { Context } from 'cordis'
|
||||
import { describe, expect, it, vi } from 'vitest'
|
||||
import LlmService, { CallId, LlmAdapter } from '@deepseek-ai/dsh-llm'
|
||||
import LlmService, { CallId, isAgentLoopRequest, LlmAdapter } from '@deepseek-ai/dsh-llm'
|
||||
import type { FinishReason, GenerateOptions, StreamChunk } from '@deepseek-ai/dsh-llm'
|
||||
import SessionStore, { SessionId } from '@deepseek-ai/dsh-session'
|
||||
import { SessionTitleProviderId } from '@deepseek-ai/dsh-session-title'
|
||||
@@ -139,6 +139,7 @@ describe('generateSessionTitleWithLlm', () => {
|
||||
const options = adapter.requests[0]!
|
||||
expect(Object.isFrozen(options)).toBe(true)
|
||||
expect(Object.isFrozen(options.messages)).toBe(true)
|
||||
expect(isAgentLoopRequest(options)).toBe(false)
|
||||
expect(options).toMatchObject({
|
||||
provider: 'current-route',
|
||||
model: 'current-model',
|
||||
|
||||
@@ -10,7 +10,7 @@ Only text blocks from human `user/message` events are eligible. The first eligib
|
||||
- `refresh(session, signal?)` materializes the fallback when needed, then explicitly runs the registered provider over the current eligible messages. Provider errors and caller cancellation reject; cancellation does not roll back a fallback append already entering durability.
|
||||
- `register(provider)` installs the sole optional provider and returns its awaitable Cordis effect disposer. A second registration throws immediately; disposal aborts pending and active calls, waits for their settlement, and only then permits another provider to register.
|
||||
|
||||
Automatic work never delays the main agent response. A provider starts only after a loop-built request's exact route matches the current logged `request/header`, including when the unchanged header needs no new snapshot. Its late completion joins an open turn or uses a flushed zero-step `session-title` turn through `ctx.sessions.appendOutOfBand()`. Automatic failures warn and retain the latest title. New all-message revisions, provider disposal, session disposal, and explicit refresh abort older work, and a stale completion cannot append. Concurrent explicit refreshes reserve their order before fallback durability waits, so only the newest call may reach the provider. Service teardown cancels queued work and drains calls that ignore cancellation before unloading completes.
|
||||
Automatic work never delays the main agent response. A provider starts only after a marked loop-built request's exact route matches the current logged `request/header`, including when the unchanged header needs no new snapshot. Its late completion joins an open turn or uses a flushed zero-step `session-title` turn through `ctx.sessions.appendOutOfBand()`. Automatic failures warn and retain the latest title. New all-message revisions, provider disposal, session disposal, and explicit refresh abort older work, and a stale completion cannot append. Concurrent explicit refreshes reserve their order before fallback durability waits, so only the newest call may reach the provider. Service teardown cancels queued work and drains calls that ignore cancellation before unloading completes.
|
||||
|
||||
Forks inherit title events in their seed unchanged. The first-message cadence does not automatically retitle a child; the all-messages cadence may append a new revision after the child receives a later human prompt.
|
||||
|
||||
|
||||
@@ -6,7 +6,7 @@
|
||||
import { Context, FiberState, Service, type Fiber } from 'cordis'
|
||||
import z from 'schemastery'
|
||||
import type { Branded } from '@deepseek-ai/dsh-brand'
|
||||
import { deepFreeze } from '@deepseek-ai/dsh-llm'
|
||||
import { deepFreeze, isAgentLoopRequest } from '@deepseek-ai/dsh-llm'
|
||||
import type { GenerateOptions } from '@deepseek-ai/dsh-llm'
|
||||
import type { Session, SessionEvent } from '@deepseek-ai/dsh-session'
|
||||
import { fallbackSessionTitle, normalizeSessionTitle } from './normalize.ts'
|
||||
@@ -413,9 +413,9 @@ export class SessionTitleService extends Service {
|
||||
this.startPending(session, state, pending, route)
|
||||
}
|
||||
|
||||
/** Start unchanged-route work from the frozen loop request after its header fold is current. */
|
||||
/** Start unchanged-route work from the marked loop request after its header fold is current. */
|
||||
private onMainRequest(options: GenerateOptions): void {
|
||||
if (!this.serviceActive() || options.sessionId === undefined || !Object.isFrozen(options)) return
|
||||
if (!this.serviceActive() || options.sessionId === undefined || !isAgentLoopRequest(options)) return
|
||||
const session = this.ctx.sessions.get(options.sessionId)
|
||||
const state = session === undefined ? undefined : this.work.get(session)
|
||||
const pending = state?.pending
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
import { Context } from 'cordis'
|
||||
import { describe, expect, it, vi } from 'vitest'
|
||||
import LlmService, { deepFreeze } from '@deepseek-ai/dsh-llm'
|
||||
import LlmService, { deepFreeze, markAgentLoopRequest } from '@deepseek-ai/dsh-llm'
|
||||
import SessionStore, { SessionId } from '@deepseek-ai/dsh-session'
|
||||
import SessionTitleService, {
|
||||
SessionTitleProviderId,
|
||||
@@ -303,12 +303,12 @@ describe('SessionTitleService provider lifecycle', () => {
|
||||
const second = appendHumanPrompt(session, 'Second prompt on the same route')
|
||||
await settle()
|
||||
session.append('step/start', { turn: 2, step: 1 })
|
||||
void ctx.llm.stream(deepFreeze({
|
||||
void ctx.llm.stream(markAgentLoopRequest(deepFreeze({
|
||||
provider: 'main-route',
|
||||
model: 'chat-model',
|
||||
messages: session.deriveMessages(),
|
||||
sessionId: session.id,
|
||||
}))
|
||||
})))
|
||||
await settle()
|
||||
|
||||
expect(session.events.filter(event => event.type === 'request/header')).toHaveLength(1)
|
||||
@@ -339,9 +339,9 @@ describe('SessionTitleService provider lifecycle', () => {
|
||||
const options = { provider: 'main-route', model: 'chat-model', messages: [] }
|
||||
|
||||
void ctx.llm.stream(deepFreeze(options))
|
||||
void ctx.llm.stream(deepFreeze({ ...options, sessionId: SessionId('missing') }))
|
||||
void ctx.llm.stream(markAgentLoopRequest(deepFreeze({ ...options, sessionId: SessionId('missing') })))
|
||||
const quiet = ctx.sessions.create(SessionId('quiet'))
|
||||
void ctx.llm.stream(deepFreeze({ ...options, sessionId: quiet.id }))
|
||||
void ctx.llm.stream(markAgentLoopRequest(deepFreeze({ ...options, sessionId: quiet.id })))
|
||||
const pending = ctx.sessions.create(SessionId('unmatched-boundary'))
|
||||
pending.append('turn/start', {
|
||||
turn: 1,
|
||||
@@ -349,7 +349,7 @@ describe('SessionTitleService provider lifecycle', () => {
|
||||
})
|
||||
appendHumanPrompt(pending, 'Wait for a matching request boundary')
|
||||
await settle()
|
||||
void ctx.llm.stream(deepFreeze({ ...options, sessionId: pending.id }))
|
||||
void ctx.llm.stream(markAgentLoopRequest(deepFreeze({ ...options, sessionId: pending.id })))
|
||||
await settle()
|
||||
|
||||
expect(generate).not.toHaveBeenCalled()
|
||||
|
||||
@@ -39,7 +39,7 @@ Agent status (per agent):
|
||||
|
||||
Model requests (on `llm/stream`):
|
||||
|
||||
- **a loop-built request is exactly what the log reconstructs** — a frozen request with a live `sessionId` (the loop-built marker; hand-built one-shots like compaction's summarize are unfrozen and skipped) must carry frozen `messages` deep-equal to the derivation over the log prefix strictly before the in-flight step's `step/start` (rebuilt through a FRESH `Session`, so the live cache cannot vouch for itself — and boundary-correct: content logged after `step/start` legitimately belongs to the next request), and every non-content field must equal the latest logged `request/header` (see [the reconstructability Agent Note](../../../.agents/notes/implemented/architecture/2026-07-05-reconstructable-requests.md)). Registered with `prepend: true` so a short-circuiting `llm/stream` listener (the replay adapter) cannot silence it; prepend orders it against append-registered listeners only — correctness rests on the seq-bounded rebuild, never listener timing.
|
||||
- **a loop-built request is exactly what the log reconstructs** — a request carrying dsh-agent-loop's process-local identity must have a frozen envelope, live `sessionId`, frozen `messages` deep-equal to the derivation over the log prefix strictly before the in-flight step's `step/start` (rebuilt through a FRESH `Session`, so the live cache cannot vouch for itself — and boundary-correct: content logged after `step/start` legitimately belongs to the next request), and every non-content field must equal the latest logged `request/header` (see [the reconstructability Agent Note](../../../.agents/notes/implemented/architecture/2026-07-05-reconstructable-requests.md)). Frozen auxiliary calls remain outside this loop-only equation. Registered with `prepend: true` so a short-circuiting `llm/stream` listener (the replay adapter) cannot silence it; prepend orders it against append-registered listeners only — correctness rests on the seq-bounded rebuild, never listener timing.
|
||||
|
||||
On any violation it throws `InvariantError` (`code: 'INVARIANT'`).
|
||||
|
||||
@@ -61,5 +61,5 @@ None; this package neither assembles nor sends a provider request.
|
||||
|
||||
## Known Limitations and Deferred Work
|
||||
|
||||
- **The request-reconstructability assertion covers loop-built requests only** — hand-built one-shots (e.g. compaction's summarize call) carry no live `sessionId` marker and are skipped.
|
||||
- **The request-reconstructability assertion covers loop-built requests only** — hand-built and auxiliary calls carry no process-local loop marker and are skipped even when they are immutable or session-associated.
|
||||
- **Merge-extended event families get no family-specific assertions** — `compact/*` lock pairing and `hook/*` invoked/result pairing are not checked here; only the core turn/step/chunk/tool-result contract is.
|
||||
|
||||
@@ -10,7 +10,7 @@
|
||||
|
||||
import type { Context } from 'cordis'
|
||||
import { carrierKeyOf, isScopeCarrier } from '@deepseek-ai/dsh-scope'
|
||||
import { assertNever, HarnessError } from '@deepseek-ai/dsh-llm'
|
||||
import { assertNever, HarnessError, isAgentLoopRequest } from '@deepseek-ai/dsh-llm'
|
||||
import type { CallId, GenerateOptions } from '@deepseek-ai/dsh-llm'
|
||||
import type { Agent, AgentStatus } from '@deepseek-ai/dsh-agent'
|
||||
import { Session, SessionId, foldRequestHeader } from '@deepseek-ai/dsh-session'
|
||||
@@ -341,8 +341,8 @@ export function apply(ctx: Context): void {
|
||||
}, { global: true })
|
||||
|
||||
// Request-reconstruction cross-check (the reconstructability Agent Note): a
|
||||
// loop-built request — frozen envelope + live sessionId is the marker; a
|
||||
// hand-built one-shot (compaction summarize) is unfrozen and skipped — must
|
||||
// loop-built request — identified by dsh-agent-loop's process-local marker;
|
||||
// frozen auxiliary calls remain outside this loop-only equation — must
|
||||
// be EXACTLY what the session log reconstructs:
|
||||
//
|
||||
// - messages: the folded header's session prefix (messagePrefix — the
|
||||
@@ -366,7 +366,13 @@ export function apply(ctx: Context): void {
|
||||
// (cordis unshift) — which is fine: correctness rests on the seq-bounded
|
||||
// fold below, never on listener timing.
|
||||
ctx.on('llm/stream', (options: GenerateOptions, next) => {
|
||||
if (options.sessionId === undefined || !Object.isFrozen(options)) return next()
|
||||
if (!isAgentLoopRequest(options)) return next()
|
||||
if (!Object.isFrozen(options)) {
|
||||
throw new InvariantError('a loop-built request must carry a frozen envelope')
|
||||
}
|
||||
if (options.sessionId === undefined) {
|
||||
throw new InvariantError('a loop-built request must carry a sessionId')
|
||||
}
|
||||
// GenerateOptions types sessionId as Branded<'SessionId'>, which IS
|
||||
// SessionId (dsh-llm cannot import it without a cycle) — no cast needed.
|
||||
const session = ctx.sessions.get(options.sessionId)
|
||||
|
||||
@@ -1,7 +1,8 @@
|
||||
import { describe, expect, it, vi } from 'vitest'
|
||||
import { Context } from 'cordis'
|
||||
import { createScope, scopeTarget } from '@deepseek-ai/dsh-scope'
|
||||
import { CallId } from '@deepseek-ai/dsh-llm'
|
||||
import { CallId, markAgentLoopRequest } from '@deepseek-ai/dsh-llm'
|
||||
import type { GenerateOptions } from '@deepseek-ai/dsh-llm'
|
||||
import type { Agent } from '@deepseek-ai/dsh-agent'
|
||||
import SessionStore, { Session, SessionId } from '@deepseek-ai/dsh-session'
|
||||
import * as Invariants from '@deepseek-ai/dsh-invariants'
|
||||
@@ -798,9 +799,14 @@ describe('request-reconstruction cross-check (llm/stream)', () => {
|
||||
void ctx.waterfall('llm/stream', options as never, () => (async function* () {})() as never)
|
||||
}
|
||||
|
||||
/** Attach the same process-local identity as dsh-agent-loop. */
|
||||
function loopRequest(options: unknown): GenerateOptions {
|
||||
return markAgentLoopRequest(options as GenerateOptions)
|
||||
}
|
||||
|
||||
it('passes a frozen request that equals the boundary derivation + the folded header', async () => {
|
||||
const { ctx, session, boundary } = await requestSetup()
|
||||
const options = Object.freeze({ model: 'm', messages: Object.freeze(boundary), sessionId: session.id })
|
||||
const options = loopRequest(Object.freeze({ model: 'm', messages: Object.freeze(boundary), sessionId: session.id }))
|
||||
expect(() => { dispatch(ctx, options) }).not.toThrow()
|
||||
})
|
||||
|
||||
@@ -810,7 +816,7 @@ describe('request-reconstruction cross-check (llm/stream)', () => {
|
||||
// belongs to the NEXT request. A current-surface comparison would
|
||||
// false-fire here; the seq-bounded rebuild must not.
|
||||
session.append('context/message', { content: [{ type: 'text', text: '[late]' }], source: { kind: 'plugin', plugin: 'x' } }, { surfaceOp: 'append' })
|
||||
const options = Object.freeze({ model: 'm', messages: Object.freeze(boundary), sessionId: session.id })
|
||||
const options = loopRequest(Object.freeze({ model: 'm', messages: Object.freeze(boundary), sessionId: session.id }))
|
||||
expect(() => { dispatch(ctx, options) }).not.toThrow()
|
||||
})
|
||||
|
||||
@@ -819,26 +825,26 @@ describe('request-reconstruction cross-check (llm/stream)', () => {
|
||||
const prefix = { role: 'user' as const, content: [{ type: 'text' as const, text: '<system-reminder>catalog</system-reminder>' }] }
|
||||
session.append('request/header', { header: { config: { provider: 'mock', model: 'm' }, messagePrefix: [prefix] }, reason: 'change' })
|
||||
// The prefixed request matches the fold…
|
||||
const prefixed = Object.freeze({ model: 'm', messages: Object.freeze([prefix, ...boundary]), sessionId: session.id })
|
||||
const prefixed = loopRequest(Object.freeze({ model: 'm', messages: Object.freeze([prefix, ...boundary]), sessionId: session.id }))
|
||||
expect(() => { dispatch(ctx, prefixed) }).not.toThrow()
|
||||
// …a request that DROPPED the logged prefix diverges…
|
||||
const bare = Object.freeze({ model: 'm', messages: Object.freeze([...boundary]), sessionId: session.id })
|
||||
const bare = loopRequest(Object.freeze({ model: 'm', messages: Object.freeze([...boundary]), sessionId: session.id }))
|
||||
expect(() => { dispatch(ctx, bare) }).toThrow(/diverges from the boundary derivation/)
|
||||
// …and so does one that misplaced it (prefix sent after the history).
|
||||
const misplaced = Object.freeze({ model: 'm', messages: Object.freeze([...boundary, prefix]), sessionId: session.id })
|
||||
const misplaced = loopRequest(Object.freeze({ model: 'm', messages: Object.freeze([...boundary, prefix]), sessionId: session.id }))
|
||||
expect(() => { dispatch(ctx, misplaced) }).toThrow(/diverges from the boundary derivation/)
|
||||
})
|
||||
|
||||
it('rejects a frozen request whose messages diverge from the boundary derivation', async () => {
|
||||
const { ctx, session, boundary } = await requestSetup()
|
||||
const messages = [...boundary, { role: 'user', content: [{ type: 'text', text: 'phantom' }] }]
|
||||
const options = Object.freeze({ model: 'm', messages: Object.freeze(messages), sessionId: session.id })
|
||||
const options = loopRequest(Object.freeze({ model: 'm', messages: Object.freeze(messages), sessionId: session.id }))
|
||||
expect(() => { dispatch(ctx, options) }).toThrow(/diverges from the boundary derivation/)
|
||||
})
|
||||
|
||||
it('rejects a frozen request whose fields diverge from the folded header', async () => {
|
||||
const { ctx, session, boundary } = await requestSetup()
|
||||
const options = Object.freeze({ model: 'other', messages: Object.freeze(boundary), sessionId: session.id })
|
||||
const options = loopRequest(Object.freeze({ model: 'other', messages: Object.freeze(boundary), sessionId: session.id }))
|
||||
expect(() => { dispatch(ctx, options) }).toThrow(/diverges from the folded request header/)
|
||||
})
|
||||
|
||||
@@ -846,7 +852,7 @@ describe('request-reconstruction cross-check (llm/stream)', () => {
|
||||
const { ctx } = await setup()
|
||||
const session = ctx.sessions.create(SessionId('req-bare'))
|
||||
session.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
|
||||
const bare = Object.freeze({ model: 'm', messages: Object.freeze([]), sessionId: session.id })
|
||||
const bare = loopRequest(Object.freeze({ model: 'm', messages: Object.freeze([]), sessionId: session.id }))
|
||||
expect(() => { dispatch(ctx, bare) }).toThrow(/no step\/start/)
|
||||
|
||||
session.append('step/start', { turn: 1, step: 1 })
|
||||
@@ -855,7 +861,7 @@ describe('request-reconstruction cross-check (llm/stream)', () => {
|
||||
|
||||
it('rejects a frozen request carrying an unfrozen messages array', async () => {
|
||||
const { ctx, session, boundary } = await requestSetup()
|
||||
const options = Object.freeze({ model: 'm', messages: [...boundary], sessionId: session.id })
|
||||
const options = loopRequest(Object.freeze({ model: 'm', messages: [...boundary], sessionId: session.id }))
|
||||
expect(() => { dispatch(ctx, options) }).toThrow(/frozen messages array/)
|
||||
})
|
||||
|
||||
@@ -866,10 +872,32 @@ describe('request-reconstruction cross-check (llm/stream)', () => {
|
||||
expect(() => { dispatch(ctx, options) }).not.toThrow()
|
||||
})
|
||||
|
||||
it('skips frozen auxiliary requests that were not built by the loop', async () => {
|
||||
const { ctx, session } = await requestSetup()
|
||||
const options = Object.freeze({
|
||||
model: 'title-model',
|
||||
messages: Object.freeze([{ role: 'user', content: [{ type: 'text', text: 'title this session' }] }]),
|
||||
sessionId: session.id,
|
||||
})
|
||||
expect(() => { dispatch(ctx, options) }).not.toThrow()
|
||||
})
|
||||
|
||||
it('rejects a marked loop request without its frozen envelope or session identity', async () => {
|
||||
const { ctx } = await requestSetup()
|
||||
expect(() => {
|
||||
dispatch(ctx, loopRequest({ model: 'm', messages: Object.freeze([]) }))
|
||||
}).toThrow(/frozen envelope/)
|
||||
expect(() => {
|
||||
dispatch(ctx, loopRequest(Object.freeze({ model: 'm', messages: Object.freeze([]) })))
|
||||
}).toThrow(/sessionId/)
|
||||
})
|
||||
|
||||
it('skips requests without a sessionId or with an unknown session', async () => {
|
||||
const { ctx } = await requestSetup()
|
||||
expect(() => { dispatch(ctx, Object.freeze({ model: 'm', messages: Object.freeze([]) })) }).not.toThrow()
|
||||
expect(() => { dispatch(ctx, Object.freeze({ model: 'm', messages: Object.freeze([]), sessionId: SessionId('ghost') })) }).not.toThrow()
|
||||
expect(() => {
|
||||
dispatch(ctx, loopRequest(Object.freeze({ model: 'm', messages: Object.freeze([]), sessionId: SessionId('ghost') })))
|
||||
}).not.toThrow()
|
||||
})
|
||||
})
|
||||
|
||||
@@ -888,11 +916,11 @@ describe('request cross-check ordering (prepend)', () => {
|
||||
session.append('step/start', { turn: 1, step: 1 })
|
||||
session.append('request/header', { header: { config: { provider: 'mock', model: 'm' } }, reason: 'initial' })
|
||||
|
||||
const divergent = Object.freeze({
|
||||
const divergent = markAgentLoopRequest(Object.freeze({
|
||||
model: 'm',
|
||||
messages: Object.freeze([{ role: 'user', content: [{ type: 'text', text: 'phantom' }] }]),
|
||||
sessionId: session.id,
|
||||
})
|
||||
}) as unknown as GenerateOptions)
|
||||
expect(() => {
|
||||
void ctx.waterfall('llm/stream', divergent as never, () => (async function* () {})() as never)
|
||||
}).toThrow(/diverges from the boundary derivation/)
|
||||
|
||||
Reference in New Issue
Block a user