Merge branch 'master' into worktree-webheadless

.agents/notes/implemented/architecture/2026-06-18-shared-persistence-write-coordinator.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
-2026-06-18-shared-persistence-write-coordinator.md: ea9c4fb74f7c1bd68fb62efedd3e1657da96ea65
-2026-06-18-shared-persistence-write-coordinator.zh.md: 3b4dd7b762c2f39a908eabe23e5d734981b5767b
+2026-06-18-shared-persistence-write-coordinator.md: 4632351a6f39c44c9ba8af58d508d4665b9e9279
+2026-06-18-shared-persistence-write-coordinator.zh.md: 40a7144038ac0db4ca6cac651c0a3cef5de4afa9

.agents/notes/implemented/architecture/2026-06-18-shared-persistence-write-coordinator.md

@@ -23,7 +23,7 @@ The coordinator retires a session from `session/disposed`: it waits for the cont
 Five required members plus an optional lifecycle hook form the only boundary between the coordinator and storage:
 
 - `name` — backend label for the dispose-failure `AggregateError`.
-- `loadStored(id)` — read one stored prefix by id across every storage scope (every JSONL cwd bucket; SQLite's id is globally unique). Resume/load, non-mutating inspection, live adoption, and the create-collision probe share this lookup. The coordinator asserts the returned id and rejects a stored/live cwd mismatch before repair or state publication.
+- `loadStored(id)` — read one stored prefix by id across every storage scope (every JSONL project directory; SQLite's id is globally unique). Resume/load, non-mutating inspection, live adoption, and the create-collision probe share this lookup. The coordinator asserts the returned id and rejects a stored/live cwd mismatch before repair or state publication.
 - `appendBatch(meta, events, isMaterialized)` — durably append a contiguous batch, lazily materializing the session ATOMICALLY when not yet materialized (the materialize-write and the first event batch must commit together — a crash between them must not leave a materialized-but-empty session; this is why there is no separate `materialize` hook).
 - `commitRepair(meta, tornMarker, closers)` — make a crash repair durable: truncate the torn tail (iff `tornMarker !== undefined`) and append `closers`. **NOT required to be atomic** — JSONL legitimately truncates-then-appends in two fsync'd steps, SQLite does DELETE+INSERT in one transaction. Used by `load` (truncate + synthetic closers) and live-adoption (truncate only, `closers = []`).
 - `list()` — list all stored metadata.

.agents/notes/implemented/architecture/2026-06-18-shared-persistence-write-coordinator.zh.md

@@ -23,7 +23,7 @@ Status: implemented
 五个必需成员加一个可选的生命周期钩子，构成协调器与存储之间唯一的边界：
 
 - `name`——后端标签，用于 dispose 失败时的 `AggregateError`。
-- `loadStored(id)`——按 id 跨所有存储范围读取一个已存储前缀（JSONL 的所有 cwd bucket；SQLite 的 id 全局唯一）。恢复/加载、不修改状态的检查、存活会话接管与创建碰撞探测共用此查找。协调器会断言返回的 id，并在修复或发布状态之前拒绝已存储记录与存活会话的 cwd 不匹配。
+- `loadStored(id)`——按 id 跨所有存储范围读取一个已存储前缀（JSONL 的所有项目目录；SQLite 的 id 全局唯一）。恢复/加载、不修改状态的检查、存活会话接管与创建碰撞探测共用此查找。协调器会断言返回的 id，并在修复或发布状态之前拒绝已存储记录与存活会话的 cwd 不匹配。
 - `appendBatch(meta, events, isMaterialized)`——持久追加一个连续批次，在尚未物化时原子地惰性物化会话（物化写入与首批事件必须一起提交——崩溃不得留下一个已物化但为空的会话；这就是为什么没有单独的 `materialize` 钩子）。
 - `commitRepair(meta, tornMarker, closers)`——使崩溃修复持久化：截断损坏的尾部（当且仅当 `tornMarker !== undefined`）并追加 `closers`。**不要求原子性**——JSONL 合理地分两步 fsync（先截断再追加），SQLite 在一个事务中完成 DELETE+INSERT。用于 `load`（截断 + 合成 closers）和 live-adoption（仅截断，`closers = []`）。
 - `list()`——列出所有已存储的元数据。

.agents/notes/implemented/architecture/2026-06-21-bounded-llm-request-recovery.md

@@ -42,7 +42,7 @@ The agent loop keeps `RequestError` as that exact error object and passes `LlmFa
 
 Adapters extract structured facts before falling back to message inspection. They validate HTTP status, parse `Retry-After` seconds or dates into a positive finite millisecond delay, brand the provider request id when exposed, and distinguish their own timeout from the caller's abort. Provider-specific codes and messages may refine a mapping, but no recovery listener parses them.
 
-The initial shared transient-code set is intentionally small: the adapters' existing `RATE_LIMIT` and `SERVER` mappings plus explicit `TIMEOUT` and `TRANSPORT` codes for the two missing remote-failure families. Authentication, quota, invalid request, context overflow, protocol, abort, and unknown failures keep distinct stable codes and are not transient by default. Adding a code requires adapter fixtures and a documented policy decision; it does not require expanding a second failure-class enum.
+The shared transient-code set is intentionally small: adapter mappings for `RATE_LIMIT` and `SERVER`, explicit `TIMEOUT` and `TRANSPORT` codes for remote failures, and `EMPTY_RESPONSE` for a completed provider response with no content blocks. Both adapters classify the last case as an error finish; see [empty model responses are retryable](../bug-fix/2026-07-24-empty-model-response-is-retryable.md). Authentication, quota, invalid request, context overflow, protocol, abort, and unknown failures keep distinct stable codes and are not transient by default. Adding a code requires adapter fixtures and a documented policy decision; it does not require expanding a second failure-class enum.
 
 ### Put retry policy on the existing failed-step seam
 
@@ -62,7 +62,7 @@ interface Config {
 }
 ```
 
-The defaults are two transient retries, a 500 millisecond initial delay, a 10 second delay cap, 10 percent jitter, and the four transient codes above. The count and delay bounds match the conservative edge of the inspected implementations: [OpenCode uses two request retries with 500 ms/10 s bounds](https://github.com/anomalyco/opencode/blob/9976269ab1accfc9f9dc98a4a688c516934de422/%70ackages/llm/src/route/executor.ts#L36-L39), [Pi separates three agent-level retries from provider retries and defaults provider retries to zero](https://github.com/earendil-works/pi/blob/3da591ab74ab9ab407e72ed882600b2c851fae21/%70ackages/coding-agent/docs/settings.md#L139-L147), and [Codex uses finite request/stream budgets plus a five-minute idle timeout](https://github.com/openai/codex/blob/0fb559f0f6e231a88ac02ea002d3ecd248e2b515/codex-rs/model-provider-info/src/lib.rs#L25-L33). Ten percent follows [Codex's bounded jitter](https://github.com/openai/codex/blob/0fb559f0f6e231a88ac02ea002d3ecd248e2b515/codex-rs/codex-client/src/retry.rs#L40-L47). Two retries mean at most three provider requests when no other recovery policy applies. `maxTransientRetries` is a non-negative integer, delays are positive finite numbers with `initialDelayMs <= maxDelayMs`, `jitterRatio` is in `[0, 1]`, and codes are non-empty and unique. These are Cordis config fields rather than hidden constants so deployments can choose different cost and latency budgets.
+The defaults are two transient retries, a 500 millisecond initial delay, a 10 second delay cap, 10 percent jitter, and the five transient codes above (`RATE_LIMIT`, `SERVER`, `TIMEOUT`, `TRANSPORT`, and `EMPTY_RESPONSE`). The count and delay bounds match the conservative edge of the inspected implementations: [OpenCode uses two request retries with 500 ms/10 s bounds](https://github.com/anomalyco/opencode/blob/9976269ab1accfc9f9dc98a4a688c516934de422/%70ackages/llm/src/route/executor.ts#L36-L39), [Pi separates three agent-level retries from provider retries and defaults provider retries to zero](https://github.com/earendil-works/pi/blob/3da591ab74ab9ab407e72ed882600b2c851fae21/%70ackages/coding-agent/docs/settings.md#L139-L147), and [Codex uses finite request/stream budgets plus a five-minute idle timeout](https://github.com/openai/codex/blob/0fb559f0f6e231a88ac02ea002d3ecd248e2b515/codex-rs/model-provider-info/src/lib.rs#L25-L33). Ten percent follows [Codex's bounded jitter](https://github.com/openai/codex/blob/0fb559f0f6e231a88ac02ea002d3ecd248e2b515/codex-rs/codex-client/src/retry.rs#L40-L47). Two retries mean at most three provider requests when no other recovery policy applies. `maxTransientRetries` is a non-negative integer, delays are positive finite numbers with `initialDelayMs <= maxDelayMs`, `jitterRatio` is in `[0, 1]`, and codes are non-empty and unique. These are Cordis config fields rather than hidden constants so deployments can choose different cost and latency budgets.
 
 For an eligible failure with budget remaining, the one-based transient retry count uses bounded exponential backoff. A valid `providerRetryAfterMs` replaces exponential backoff only when it does not exceed `maxDelayMs`; a longer provider delay causes delegation instead of an earlier retry that violates the provider instruction. Local backoff multiplies by an injected random factor in `[1 - jitterRatio, 1 + jitterRatio]` and clamps the final value to `maxDelayMs`; provider delay is not jittered.
 
@@ -124,7 +124,7 @@ If recovery is exhausted, the final failure is stored once on `turn/end.reason`
 - Pure unit tests cover transient-code selection, exponential backoff and jitter bounds, valid and over-cap `Retry-After`, exhausted budgets, deterministic timer/random seams, and abort during backoff.
 - Real agent-loop tests cover failure before chunks, partial chunks then failure, thrown and in-band failures, retry to success in a new step, exhaustion to structured `turn/end.reason`, and composition with `dsh-compact-basic` context-overflow recovery.
 - The partial-chunk integration test proves failed chunks remain attributed to the failed step, no assistant message or tool side effect is committed for that step, and the successful retry has distinct provenance.
-- The plugin-owned `llm/retry` event is non-surface, survives JSONL and SQLite round trips, is ignored by message derivation, and drives TUI retraction plus durable discarded-attempt markers in append-only ACP and stdio streams. Keyless snapshots cover scheduling, cancellation, success, and exhaustion.
+- The plugin-owned `llm/retry` event is non-surface, survives JSONL and SQLite round trips, is ignored by message derivation, and drives TUI retraction plus scheduled-retry rendering. Keyless snapshots cover scheduling, cancellation, success, and exhaustion; ACP automation snapshots confirm that a discarded attempt stays off the wire while the recovered reply is emitted.
 - Idle-watchdog tests prove the stable signal is rearmed only while `next()` is outstanding, disarmed during consumer think time and in `finally`, and classified separately from a total-call deadline and an earlier caller abort; adapter tests prove the signal stops the underlying request rather than merely detaching it.
 - Direct `ctx.llm.stream()` callers remain single-attempt and receive the same structured failure facts.
 

.agents/notes/implemented/architecture/2026-07-05-windows-jsonl-durable-publish.md

@@ -12,9 +12,9 @@ Windows has atomic namespace operations, but Node does not expose a POSIX-equiva
 
 The JSONL backend forks inside `materialize()` before any namespace mutation. Shared code computes the session directory, final log path, and encoded header plus initial event batch; POSIX and Windows then run separate publication protocols.
 
-POSIX keeps the existing protocol: create the root and cwd bucket with parent directory fsyncs, write and fsync a temp file, publish with `link()` so an existing final log is never overwritten, fsync the bucket directory, then remove the redundant temp hard link.
+POSIX keeps the existing protocol: create the root, project directory, and session directory with parent directory fsyncs, write and fsync a temp file, publish with `link()` so an existing final log is never overwritten, fsync the session directory, then remove the redundant temp hard link.
 
-Windows creates missing directories through a durable staging publish: create a random sibling directory, then publish it to the final directory name with `MoveFileExW(..., MOVEFILE_WRITE_THROUGH)` without `MOVEFILE_REPLACE_EXISTING` or `MOVEFILE_COPY_ALLOWED`. File materialization writes and fsyncs the temp log, then publishes that temp file to the final path with the same write-through `MoveFileExW` call and no replacement. `koffi` is the minimal Win32 bridge for this API surface; its install script is allowed in `pnpm-workspace.yaml` because the package ships the native loader and prebuilt platform modules.
+Windows creates missing directories through a durable staging publish: create a random sibling directory under the constant `.dsh-mkdir-` prefix, independent of the target basename, then publish it to the final directory name with `MoveFileExW(..., MOVEFILE_WRITE_THROUGH)` without `MOVEFILE_REPLACE_EXISTING` or `MOVEFILE_COPY_ALLOWED`. File materialization writes and fsyncs the temp log, then publishes that temp file to the final path with the same write-through `MoveFileExW` call and no replacement. `koffi` is the minimal Win32 bridge for this API surface; its install script is allowed in `pnpm-workspace.yaml` because the package ships the native loader and prebuilt platform modules.
 
 ## Alternatives considered
 
@@ -28,6 +28,6 @@ Windows creates missing directories through a durable staging publish: create a
 
 The backend keeps one external contract across platforms: first append either publishes a complete log at the final name or fails without overwriting an existing log. The platform split is an implementation detail; `SessionPersistence` APIs and the logical JSONL record format do not change. The later [Zstandard encoding decision](2026-07-19-zstandard-jsonl-session-logs.md) applies before either platform publishes the opaque bytes.
 
-Windows tests exercise the real Win32 publish path on native Windows. Power-loss behavior remains an API-contract property rather than something unit tests can prove; the testable invariants are that directory fsync is not called on Windows materialization, final-path collisions fail, temp logs are fsync'd before publication, and the resulting log loads normally.
+Windows tests exercise the real Win32 publish path on native Windows. Power-loss behavior remains an API-contract property rather than something unit tests can prove; the testable invariants are that directory fsync is not called on Windows materialization, final-path collisions fail, maximum-length target components remain materializable, temp logs are fsync'd before publication, and the resulting log loads normally.
 
 Append and repair still use ordinary file-handle fsyncs on both platforms. A failed append closes its append-only handle, reopens the log read/write, truncates it to the pre-append size, and fsyncs the rollback because Windows rejects `ftruncate` on append-only handles.

.agents/notes/implemented/architecture/2026-07-24-project-session-directories.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
+2026-07-24-project-session-directories.md: 0aa3f513d5a1bb3e44cf33a0ae1eb791ee3a46c2
+2026-07-24-project-session-directories.zh.md: 3d8d33fa9fddad010ab319ac4e1f873b69b4e1dd

.agents/notes/implemented/architecture/2026-07-24-project-session-directories.md

@@ -0,0 +1,52 @@
+# Agent Note: Project-grouped session directories
+
+Status: implemented
+
+English | [中文](2026-07-24-project-session-directories.zh.md)
+
+## Problem
+
+A persistence root may be local to one project, shared by several projects, temporary, or centralized. The hashed cwd buckets kept all deployments functional but made a shared root difficult to navigate because a developer could not recognize a project from its directory name.
+
+Each JSONL session also occupied one file directly inside the project bucket. That shape had no ownership directory for additional session artifacts such as metadata, attachments, spill files, or coordination state.
+
+## Decision
+
+The JSONL backend stores sessions under a readable project key and gives every session its own directory:
+
+```text
+<configured-root>/
+  --<normalized-cwd>--/
+    <encoded-session-id>/
+      session.jsonl.zstd
+```
+
+Raw mode uses `session.jsonl`, and sessions without a cwd use `_no-cwd`. Filesystem and drive separators become `-`, unsafe code units use `~XXXX`, and the readable name is bounded to keep the component within filesystem limits.
+
+The project key intentionally has no hash suffix. This follows the common human-readable convention used by coding agents and keeps the normalized project path as the complete directory name. The normalization is lossy: paths such as `/a/b-c` and `/a-b/c`, or long paths with the same retained prefix, share one project directory. Their distinct session ids still select separate session directories; reuse of the same session id remains a storage collision and is rejected.
+
+Case-insensitive filesystems can also make differently cased project keys refer to one physical directory. Identity validation accepts such an alternate spelling only when filesystem canonicalization resolves the discovered and expected paths to the same transcript. A different canonical path remains corruption, so case aliases do not weaken the same-id collision check on case-sensitive stores.
+
+The configured root remains a deployment choice. The layout neither selects a global root nor requires projects to share one. When a deployment does centralize storage, project paths remain recognizable; a project-local root uses the same deterministic structure.
+
+The encoded session id names an ownership directory rather than the transcript itself. `SessionPersistence.locate()` continues to return the fixed transcript path, preserving hook `transcript_path` and `DSH_SESSION_JSONL` semantics. Discovery ignores other entries inside the session directory so the backend can add session-owned artifacts without another layout change.
+
+Lazy materialization remains tied to the transcript: `create()` performs no filesystem I/O, and the first append creates the project/session directories before collision-safe transcript publication. Empty directories are not listed as sessions. The backend rejects flat `<project>/<id>.jsonl*` artifacts with an explicit layout error; the pre-release format provides no automatic data migration.
+
+## Alternatives considered
+
+**Keep opaque cwd hashes.** This preserved short names but defeated the requested navigation by project path when several projects share a persistence root.
+
+**Put session files directly in each project directory.** This matched Claude Code and pi's basic file organization but left no session-level ownership boundary for future artifacts.
+
+**Add a collision-resistant hash suffix.** This distinguishes paths whose normalized forms collide, but makes the directory name more than the normalized project path. The chosen convention accepts lossy project grouping in exchange for the simpler, recognizable name.
+
+**Mandate a centralized root.** Rejected because storage placement belongs to deployment configuration. Project grouping is useful when roots are shared and harmless when they are not.
+
+**Load both flat and directory layouts.** Rejected under the pre-release no-compatibility stance. One accepted layout keeps identity checks and discovery deterministic.
+
+## Consequences
+
+Shared stores can be navigated by recognizable project names, while local and custom roots keep their existing configuration freedom. Every session has a directory available for future backend-owned artifacts, and existing transcript consumers still receive a file path.
+
+Project directory names are longer than the former 12-hex cwd hashes. Very long paths show only a bounded prefix. Moving a project usually selects a different directory, but distinct cwd strings that normalize to the same name share one project directory by design.

.agents/notes/implemented/architecture/2026-07-24-project-session-directories.zh.md

@@ -0,0 +1,52 @@
+# Agent Note: 按项目分组的会话目录
+
+Status: implemented
+
+[English](2026-07-24-project-session-directories.md) | 中文
+
+## 问题
+
+持久化根目录可以只供一个项目使用，也可以由多个项目共享，还可以是临时目录或集中式目录。对 cwd 进行哈希得到的分桶目录能适用于所有这些部署方式，但开发者无法从目录名辨认项目，因此共享根目录难以浏览。
+
+每个 JSONL 会话也直接以单个文件的形式放在项目分桶目录中。这种布局没有为元数据、附件、溢写文件或协调状态等其他会话产物提供归属目录。
+
+## 决策
+
+JSONL 后端按可读的项目键存储会话，并为每个会话提供独立目录：
+
+```text
+<configured-root>/
+  --<normalized-cwd>--/
+    <encoded-session-id>/
+      session.jsonl.zstd
+```
+
+原始模式使用 `session.jsonl`，没有 cwd 的会话使用 `_no-cwd`。文件系统路径分隔符和驱动器分隔符会转换为 `-`，不安全的代码单元使用 `~XXXX`，可读名称则限制长度，以确保目录项不超过文件系统限制。
+
+项目键有意不带哈希后缀。这遵循 coding agent（编码智能体）常用的易读约定，使规范化后的项目路径本身就是完整的目录名。规范化过程有损：`/a/b-c` 与 `/a-b/c` 等路径，或者保留前缀相同的长路径，会共用同一个项目目录。不同的会话 id 仍会选择不同的会话目录；复用相同的会话 id 仍构成存储冲突，系统会予以拒绝。
+
+在不区分大小写的文件系统上，大小写不同的项目键也可能指向同一个物理目录。只有当文件系统路径规范化将发现路径和预期路径解析为同一个 transcript（文本记录）时，身份验证才接受这种拼写变体。规范化后的路径如果不同，仍视为存储损坏，因此大小写别名不会让区分大小写的存储放宽同一 id 的冲突检查。
+
+根目录由部署配置决定。这种布局既不选择全局根目录，也不要求项目共享根目录。部署选择集中存储时，目录名仍能让项目路径易于辨认；使用项目本地根目录时，也采用同样的确定性结构。
+
+编码后的会话 id 用于命名归属目录，而不是 transcript 文件本身。`SessionPersistence.locate()` 仍返回固定的 transcript 路径，从而保持钩子 `transcript_path` 和 `DSH_SESSION_JSONL` 的语义不变。发现过程会忽略会话目录中的其他条目，因此后端以后添加会话自有产物时无需再次改变布局。
+
+延迟物化仍以 transcript 为界：`create()` 不执行文件系统 I/O，首次追加会先创建项目目录和会话目录，再以无冲突方式发布 transcript。空目录不会被列为会话。后端会显式报告布局错误并拒绝扁平的 `<project>/<id>.jsonl*` 产物；预发布格式不提供自动数据迁移。
+
+## 考虑过的替代方案
+
+**保留不透明的 cwd 哈希。** 这可以保持目录名简短，但当多个项目共享一个持久化根目录时，无法满足按项目路径浏览的需求。
+
+**把会话文件直接放入各项目目录。** 这与 Claude Code 和 pi 的基本文件组织一致，但没有为未来产物提供会话级归属边界。
+
+**添加防冲突的哈希后缀。** 这种方式能区分规范化形式相同的路径，但会使目录名不再只是规范化后的项目路径。所选约定接受有损的项目分组，以换取更简单、易于辨认的名称。
+
+**强制使用集中式根目录。** 不予采纳，因为存储位置属于部署配置。项目分组在根目录共享时有用，在不共享时也没有负面影响。
+
+**同时加载扁平布局和目录布局。** 按照预发布阶段不提供兼容性的原则，不予采纳。只接受一种布局，可以让身份检查和发现过程保持确定性。
+
+## 后果
+
+共享存储可以通过易于辨认的项目名进行浏览，本地根目录和自定义根目录则继续保有现有的配置自由。每个会话都有一个可供后端未来存放自有产物的目录，而现有 transcript 消费方仍会收到文件路径。
+
+项目目录名比原先由 12 个十六进制字符组成的 cwd 哈希更长。路径很长时，目录名只显示长度受限的前缀。移动项目通常会选择不同的目录，但按设计，不同的 cwd 字符串如果规范化成相同名称，就会共用同一个项目目录。

.agents/notes/implemented/bug-fix/2026-07-20-jsonl-storage-identity.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
-2026-07-20-jsonl-storage-identity.md: 1ada16791f411a54fbcf9271c7d7963223bbe683
-2026-07-20-jsonl-storage-identity.zh.md: 8027c51dbf6c7d01463b7851d859a40890bf03e1
+2026-07-20-jsonl-storage-identity.md: 1079eb700c819951dbb81e99376c0b71e3e84617
+2026-07-20-jsonl-storage-identity.zh.md: d7ba5c646a7adaaa0ebd60fac7b9c2f030361ff9

.agents/notes/implemented/bug-fix/2026-07-20-jsonl-storage-identity.md

@@ -6,11 +6,11 @@ English | [中文](2026-07-20-jsonl-storage-identity.zh.md)
 
 ## Problem
 
-JSONL lookup selects a physical log from the requested session id across cwd buckets, while the parsed `SessionHeader` supplies the metadata used by later repair and append operations. Without binding those two facts, a log selected for session A can declare session B's id or cwd and redirect a repair or later append to B's path. The bucket scan also needs a defined result when the same encoded id exists in more than one bucket. SQLite does not share this ambiguity because its primary-key query binds metadata and events to the requested id.
+JSONL lookup selects a physical log from the requested session id across project directories, while the parsed `SessionHeader` supplies the metadata used by later repair and append operations. Without binding those two facts, a log selected for session A can declare session B's id or cwd and redirect a repair or later append to B's path. The project scan also needs a defined result when the same encoded id exists in more than one project directory. SQLite does not share this ambiguity because its primary-key query binds metadata and events to the requested id.
 
 ## Decision
 
-`loadStored(id)` is the coordinator's single stored-prefix lookup. The JSONL backend scans every cwd bucket, requires at most one matching encoded filename, parses that file, then validates both `header.id === id` and `selectedPath === logPath(root, header.cwd, header.id)` before returning metadata. `list()` applies the same path validation and rejects duplicate ids across buckets.
+`loadStored(id)` is the coordinator's single stored-prefix lookup. The JSONL backend scans every project directory, requires at most one matching encoded session directory with a transcript, parses that file, then validates `header.id === id` and that the selected path either equals `logPath(root, header.cwd, header.id)` or filesystem canonicalization resolves both spellings to the same transcript. `list()` applies the same path validation and rejects duplicate ids across project directories.
 
 The coordinator independently asserts the returned id and compares the stored cwd with a live session's cwd before repair, state publication, or suffix persistence. It keeps a detached copy of validated metadata; JSONL append and repair derive their path from that copy. The `PersistenceBackend<TornMarker>` interface therefore needs neither a scope-specific live lookup nor a storage-locator type.
 
@@ -18,7 +18,7 @@ An existing configured JSONL root must be a readable directory when the plugin l
 
 ## Alternatives considered
 
-**Flatten storage by session id.** A flat namespace makes duplicate publication collide on one path, but path validation and duplicate rejection close the identity defect without changing the project-grouped cwd layout or its consumers.
+**Flatten storage by session id.** A flat namespace makes duplicate publication collide on one path, but path validation and duplicate rejection close the identity defect without making the check depend on a flat global namespace.
 
 **Carry an opaque storage locator through the coordinator.** A locator binds JSONL mutations directly to a selected path, but JSONL can reproduce that path from metadata it has already validated. Adding another generic and argument to SQLite, test backends, append, and repair makes every implementation carry a concept only the file backend needs.
 
@@ -26,4 +26,4 @@ An existing configured JSONL root must be a readable directory when the plugin l
 
 ## Consequences
 
-Mismatched, misplaced, and duplicate JSONL logs fail before repair or coordinator state mutation. The cwd-bucket format stays unchanged and needs no migration. Lookup remains proportional to the number of buckets, and one-live-writer ownership remains an explicit limitation. Coordinator and JSONL tests pin rejection before repair, unchanged bytes for both affected logs, path validation during listing, duplicate-id rejection, cwd collision handling, and load-time root validation.
+Mismatched, misplaced, and duplicate JSONL logs fail before repair or coordinator state mutation. Lookup remains proportional to the number of project directories, and one-live-writer ownership remains an explicit limitation. Coordinator and JSONL tests pin rejection before repair, unchanged bytes for both affected logs, path validation during listing, duplicate-id rejection, normalized-project collisions and case aliases, and load-time root validation.

.agents/notes/implemented/bug-fix/2026-07-20-jsonl-storage-identity.zh.md

@@ -6,11 +6,11 @@ Status: implemented
 
 ## 问题
 
-JSONL 查找会根据请求的会话 id 在各个 cwd 分桶目录中选出物理日志，而解析得到的 `SessionHeader` 会提供后续修复和追加操作使用的元数据。如果这两个事实没有绑定，为会话 A 选中的日志就能声明会话 B 的 id 或 cwd，并将修复或后续追加重定向到 B 的路径。当同一个编码后 id 出现在多个分桶目录中时，分桶扫描也必须给出确定的结果。SQLite 不存在这种歧义，因为主键查询会将元数据和事件绑定到请求的 id。
+JSONL 查找会根据请求的会话 id 在各个项目目录中选出物理日志，而解析得到的 `SessionHeader` 会提供后续修复和追加操作使用的元数据。如果这两个事实没有绑定，为会话 A 选中的日志就能声明会话 B 的 id 或 cwd，并将修复或后续追加重定向到 B 的路径。当同一个编码后 id 出现在多个项目目录中时，项目扫描也必须给出确定的结果。SQLite 不存在这种歧义，因为主键查询会将元数据和事件绑定到请求的 id。
 
 ## 决策
 
-`loadStored(id)` 是协调器唯一的已存前缀查找操作。JSONL 后端扫描所有 cwd 分桶目录，要求匹配编码文件名的日志至多有一个，解析该文件，然后在返回元数据前同时验证 `header.id === id` 和 `selectedPath === logPath(root, header.cwd, header.id)`。`list()` 执行相同的路径验证，并拒绝跨分桶目录重复的 id。
+`loadStored(id)` 是协调器唯一的已存前缀查找操作。JSONL 后端扫描所有项目目录，要求名称与该 id 的编码值匹配且其中包含 transcript（文本记录）的会话目录至多有一个，解析其中的 transcript，然后验证 `header.id === id`，并验证选定路径要么等于 `logPath(root, header.cwd, header.id)`，要么经文件系统路径规范化后，两种写法解析为同一份 transcript。`list()` 执行相同的路径验证，并拒绝跨项目目录重复的 id。
 
 协调器会独立断言返回的 id，并在修复、发布状态或持久化后缀之前比较已存 cwd 和活动会话的 cwd。协调器保留一份已验证元数据的独立副本；JSONL 的追加和修复操作根据该副本派生路径。因此，`PersistenceBackend<TornMarker>` 接口既不需要限定范围的活动会话查找，也不需要存储定位器类型。
 
@@ -18,7 +18,7 @@ JSONL 查找会根据请求的会话 id 在各个 cwd 分桶目录中选出物
 
 ## 考虑过的替代方案
 
-**按会话 id 扁平化存储。** 扁平命名空间会让重复发布在同一路径上冲突，但路径验证和重复项拒绝无需改变按项目分组的 cwd 布局及其消费方，也能消除身份缺陷。
+**按会话 id 扁平化存储。** 扁平命名空间会让重复发布在同一路径上冲突，但路径验证和重复项拒绝无需让检查依赖扁平的全局命名空间，也能消除身份缺陷。
 
 **通过协调器传递不透明存储定位器。** 定位器可以将 JSONL 变更直接绑定到选定路径，但 JSONL 可以根据已经验证的元数据重新得到该路径。为 SQLite、测试后端、追加和修复操作增加一个泛型和参数，会让每个实现都承担只有文件后端需要的概念。
 
@@ -26,4 +26,4 @@ JSONL 查找会根据请求的会话 id 在各个 cwd 分桶目录中选出物
 
 ## 后果
 
-JSONL 日志的身份不匹配、位置错误和重复会在修复或协调器状态变更前失败。cwd 分桶格式保持不变，无需迁移。查找开销仍与分桶目录数量成正比，单一活动写入方的所有权仍是明确限制。协调器和 JSONL 测试固定了修复前拒绝、两个受影响日志的字节均保持不变、列出时的路径验证、重复 id 拒绝、cwd 冲突处理以及加载时的根目录验证。
+JSONL 日志的身份不匹配、位置错误和重复会在修复或协调器状态变更前失败。查找开销仍与项目目录数量成正比，单一活动写入方的所有权仍是明确限制。协调器和 JSONL 测试固定了修复前拒绝、两个受影响日志的字节均保持不变、列出时的路径验证、重复 id 拒绝、项目路径规范化冲突与大小写别名，以及加载时的根目录验证。

.agents/notes/implemented/bug-fix/2026-07-24-empty-model-response-is-retryable.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
+2026-07-24-empty-model-response-is-retryable.md: f4a6373178efd5ca1ba5882fb2aaf97dffb2526b
+2026-07-24-empty-model-response-is-retryable.zh.md: 4c3afe44140c029d274f34ade97803b958c6d669

.agents/notes/implemented/bug-fix/2026-07-24-empty-model-response-is-retryable.md

@@ -0,0 +1,36 @@
+# Agent Note: Empty model completions are retryable EMPTY_RESPONSE failures
+
+Status: implemented
+
+English | [中文](2026-07-24-empty-model-response-is-retryable.zh.md)
+
+## Problem
+
+Providers occasionally return a degenerate completion: a well-formed stream that carries a terminal `stop` finish and zero content blocks — no text, no reasoning, no tool calls. If an adapter maps this shape to a successful `{kind: 'stop'}` finish, the loop logs an empty `assistant/message` and ends the turn as `completed`. Retry never runs, no failure reaches the caller, and a driver such as goal-session consumes a round without progress.
+
+## Decision
+
+An adapter classifies a completed empty response as a provider-boundary failure, and retry policy treats it as transient:
+
+- `dsh-llm` exports the canonical code `EMPTY_RESPONSE_CODE` (`'EMPTY_RESPONSE'`) beside `CONTEXT_WINDOW_EXCEEDED_CODE`/`QUOTA_EXCEEDED_CODE`.
+- `dsh-llm-pi-ai` (`mapStopReason`): a terminal `stop` whose assistant message has no content blocks becomes a `finish {kind: 'error'}` with that code. Context-overflow detection still wins where it applies (it is checked first and is the more actionable classification).
+- `dsh-llm-deepseek` (`translate`): at `[DONE]`, a `stop` (or absent) finish with no opened blocks becomes the same error finish. Reasoning-only streams count as content and stay successful.
+- `dsh-llm-retry` adds `EMPTY_RESPONSE` to `DEFAULT_RETRYABLE_CODES`: the attempt produced nothing durable, so repeating it is safe; deployments can still remove it via `retryableCodes`.
+
+Detection is scoped to `stop` finishes only. `max-tokens` with empty content keeps its existing meaning (pi-ai already normalizes the zero-output overflow case), `tool-calls` cannot be block-empty in practice, and error/aborted finishes already fail.
+
+The classification uses the existing loop machinery — `finishError` → `agent/request-error` → `dsh-llm-retry` — and keeps `agent-loop` provider-neutral. Exhausting the retry budget ends the turn with an explicit `EMPTY_RESPONSE` failure instead of an empty success.
+
+## Alternatives considered
+
+**Detect in the loop or `BlockAssembler`.** One shared implementation, but it moves provider-response judgment into the loop, against "plugins, not loop changes", and the assembler is a pure assembly algorithm. The adapter is where wire facts become harness classification, with the overflow reclassification as exact precedent.
+
+**A stream-transform plugin on the `llm/stream` waterfall.** Provider-neutral and one implementation, but it adds a package plus wiring for what is a boundary fact each adapter can state in a few lines, and default-on behavior would still require touching every bundle.
+
+**Treat whitespace-only or reasoning-only responses as empty too.** Rejected as overreach: those carry model-produced content, and misclassifying a legitimate (if useless) response as a transport-class failure risks retry loops on models that intentionally stop after reasoning. The scope is exactly "zero content blocks".
+
+## Consequences
+
+- A transiently misbehaving provider consumes a bounded retry instead of a turn with no output; a persistently empty model surfaces an actionable `EMPTY_RESPONSE` turn failure.
+- A model that genuinely intends to say nothing (rare, but possible after a tool result) is retried and, if consistently empty, fails the turn. This trade was accepted deliberately: an empty assistant message is indistinguishable from the provider defect and has no value to the user.
+- The `empty-response-retry` ACP snapshot (an authored keyless scenario with a deterministic 1 ms zero-jitter retry overlay, `examples/acp-agent/retry.cordis.yml`) pins the product-visible behavior: a durable `llm/retry` event, no ACP output for the discarded attempt, the recovered reply, and a clean completed turn.

.agents/notes/implemented/bug-fix/2026-07-24-empty-model-response-is-retryable.zh.md

@@ -0,0 +1,36 @@
+# Agent Note: Empty model completions are retryable EMPTY_RESPONSE failures
+
+Status: implemented
+
+[English](2026-07-24-empty-model-response-is-retryable.md) | 中文
+
+## Problem
+
+提供方偶尔会返回一种退化的 completion：流本身格式完好，携带一个终止性的 `stop` 结束，却没有任何内容块——没有文本、没有 reasoning（推理）、没有工具调用。如果适配器把这种形态映射为成功的 `{kind: 'stop'}` 结束，主循环就会记录一条空的 `assistant/message`，并把该轮次以 `completed` 结束。系统不会重试，失败也不会向调用方暴露，而像 goal-session 这样的驱动方会消耗一个轮次，却没有取得任何进展。
+
+## Decision
+
+由适配器把「已完成但为空」的响应归类为一次提供方边界失败，重试策略则将其视为瞬时性问题：
+
+- `dsh-llm` 在 `CONTEXT_WINDOW_EXCEEDED_CODE`/`QUOTA_EXCEEDED_CODE` 之外，导出规范代码 `EMPTY_RESPONSE_CODE`（`'EMPTY_RESPONSE'`）。
+- `dsh-llm-pi-ai`（`mapStopReason`）：当终止性 `stop` 所对应的 assistant 消息没有内容块时，它会变成一个携带该代码的 `finish {kind: 'error'}`。上下文溢出检测在其适用场景中仍然优先（它先被检查，也是更具可操作性的归类）。
+- `dsh-llm-deepseek`（`translate`）：在 `[DONE]` 处，若 `stop`（或缺失）结束且没有打开过任何块，则同样变成该错误结束。仅含 reasoning 的流算作有内容，仍视为成功。
+- `dsh-llm-retry` 把 `EMPTY_RESPONSE` 加入 `DEFAULT_RETRYABLE_CODES`：这次尝试没有产生任何持久内容，因此重复它是安全的；部署方仍可通过 `retryableCodes` 将其移除。
+
+检测仅限于 `stop` 结束。内容为空的 `max-tokens` 保持其既有含义（pi-ai 已经把零输出的溢出场景归一化处理），`tool-calls` 在实践中不可能是空块，而 error／aborted 结束本身已经算失败。
+
+这套归类使用既有的主循环机制——`finishError` → `agent/request-error` → `dsh-llm-retry`——并让 `agent-loop` 保持提供方无关。重试预算耗尽时，该轮次会以显式的 `EMPTY_RESPONSE` 失败结束，而不是在没有内容的情况下成功结束。
+
+## Alternatives considered
+
+**在主循环或 `BlockAssembler` 中检测。** 只需一份共享实现，但这会把对提供方响应的判断挪进主循环，违背「插件优先，而非改动主循环」，且 assembler 是纯粹的组装算法。适配器才是把协议层面的事实转化为 harness 归类的地方，而溢出重归类正是精确的先例。
+
+**在 `llm/stream` waterfall（瀑布式事件）上做一个流转换插件。** 这种做法提供方无关且只需一份实现，但它为「每个适配器几行就能声明的边界事实」额外增加了一个包和相应接线，而且默认开启的行为仍需改动每一个 bundle。
+
+**把仅含空白或仅含 reasoning 的响应也当作空响应。** 作为过度设计予以否决：这类响应携带了模型产生的内容，把一个合法（哪怕无用）的响应误判为传输类失败，会在那些故意在 reasoning 之后停止的模型上引发重试循环。其范围严格限定为「零内容块」。
+
+## Consequences
+
+- 一个偶发异常的提供方会消耗一次有界重试，而不是一个没有输出的轮次；一个持续返回空内容的模型则会暴露为用户可据以行动的 `EMPTY_RESPONSE` 轮次失败。
+- 一个确实打算什么都不说的模型（罕见，但在一次工具结果之后有可能出现）会被重试，若始终为空，则该轮次失败。这个取舍是经过审慎权衡后接受的：一条空的 assistant 消息与提供方缺陷无法区分，且对用户毫无价值。
+- `empty-response-retry` ACP 快照（一个人工编写的无密钥场景，配有确定性的 1 ms 零抖动重试 overlay，`examples/acp-agent/retry.cordis.yml`）钉住了产品可见的行为：持久的 `llm/retry` 事件、被丢弃的尝试不产生任何 ACP 输出、恢复后的回复，以及一次干净的已完成轮次。

.agents/notes/implemented/testing/2026-06-22-subagent-snapshot-replay.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
-2026-06-22-subagent-snapshot-replay.md: 6e5e94308ed145b83160146fd9e9ef023f2dde5d
-2026-06-22-subagent-snapshot-replay.zh.md: 82bb7d0735c7dbf918941d00ee4c59498cc59085
+2026-06-22-subagent-snapshot-replay.md: 8cd7bc86e07af9ed274c18574b575b9070854e88
+2026-06-22-subagent-snapshot-replay.zh.md: eae78129405fedd03c2c579845c07c6e5694cc30

.agents/notes/implemented/testing/2026-06-22-subagent-snapshot-replay.md

@@ -11,7 +11,7 @@ The snapshot tier (`pnpm run test:snapshot`) boots the real `acp-agent` subproce
 It was built for ONE session per process, and that assumption is wired into two places:
 
 - **`dsh-llm-replay` keyed nothing.** It served the Nth `llm/stream` call the Nth recorded entry from a single global cursor. With a parent agent AND an in-process subagent both streaming on one context, the calls interleave and the single cursor hands the child the parent's script (and vice versa).
-- **The harness harvested one log.** `findSessionLog` walked the sessions root and returned the FIRST `.jsonl` it found. A subagent runs as a second `Session` with its own log in the same cwd bucket, so the child's transcript was silently dropped.
+- **The harness harvested one log.** `findSessionLog` walked the sessions root and returned the FIRST `.jsonl` it found. A subagent runs as a second `Session` with its own log, so the child's transcript was silently dropped.
 
 This was the `TODO(subagent-snapshots)` deferral recorded in the [subagent seam Agent Note](../feature/2026-06-21-subagent-capability-seam.md): the in-process backends (PR2) shipped with unit + e2e coverage, but the full-transcript snapshot tier could not express a nested-agent shape until this infrastructure landed. This Agent Note is that stacked follow-up.
 
@@ -39,7 +39,7 @@ The alternative considered and rejected was a **call-ordered merge of the parent
 
 ### 3. The harness harvests every log, primary-first
 
-`harvestSessionLogs` collects every `.jsonl` across every cwd bucket under the sessions root (the JSONL backend puts a parent and its same-cwd child in the same bucket), parses each header, and orders them primary-first: the top-level session (no `parentSession`) leads, then each child by ascending `createdAt`. `RunResult.sessionLogs` is the plural result; the spec writes each back to its fixture on record (`session.jsonl` + `session.<n>.jsonl`) and diffs each harvested log against its fixture on replay. The normalizer already accepted plural session ids and collapses any stray UUID, so no normalizer change was needed.
+`harvestSessionLogs` recursively collects every fixed `session.jsonl` transcript under the sessions root (the JSONL backend gives each parent and child its own project/session directory), parses each header, and orders them primary-first: the top-level session (no `parentSession`) leads, then each child by ascending `createdAt`. `RunResult.sessionLogs` is the plural result; the spec writes each back to its fixture on record (`session.jsonl` + `session.<n>.jsonl`) and diffs each harvested log against its fixture on replay. The normalizer already accepted plural session ids and collapses any stray UUID, so no normalizer change was needed.
 
 ### 4. Scenarios
 

.agents/notes/implemented/testing/2026-06-22-subagent-snapshot-replay.zh.md

@@ -11,7 +11,7 @@ Status: implemented
 该层最初为每个进程只有一个会话而构建，这一假设硬编码在两处：
 
 - **`dsh-llm-replay` 没有做任何键控。** 它用一个全局游标，将第 N 次 `llm/stream` 调用对应到单一录制序列的第 N 条。当父 agent（智能体）和一个进程内 subagent 在同一个上下文上同时流式输出时，调用交错，单一游标会把子 agent 的脚本发给父 agent（反之亦然）。
-- **harness 只收集一份日志。** `findSessionLog` 遍历 sessions 根目录，返回找到的第一个 `.jsonl`。subagent 作为第二个 `Session` 运行，在同一个 cwd bucket 下有自己的日志，因此子 agent 的 transcript（文本记录）被静默丢弃。
+- **harness 只收集一份日志。** `findSessionLog` 遍历 sessions 根目录，返回找到的第一个 `.jsonl`。subagent 作为第二个 `Session` 运行并拥有自己的日志，因此子 agent 的 transcript（文本记录）被静默丢弃。
 
 这就是 [subagent seam Agent Note（agent 决策记录）](../feature/2026-06-21-subagent-capability-seam.md)中通过 `TODO(subagent-snapshots)` 推迟的工作：进程内后端（PR2）落地时已有单元 + e2e 覆盖，但在这套基础设施落地前，完整 transcript 快照层无法表达嵌套 agent 形状。本 Agent Note 就是该堆叠式后续工作。
 
@@ -39,7 +39,7 @@ Status: implemented
 
 ### 3. harness 收集所有日志，主会话优先
 
-`harvestSessionLogs` 收集 sessions 根目录下每个 cwd bucket 中的所有 `.jsonl`（JSONL 后端将父会话与同 cwd 的子会话放在同一个 bucket），解析各自的 header，并按主会话优先排序：顶层会话（无 `parentSession`）在前，各子会话按 `createdAt` 升序排列。`RunResult.sessionLogs` 是复数结果；spec 在录制时将每份日志写回对应 fixture（`session.jsonl` + `session.<n>.jsonl`），在回放时将每份收集到的日志与其 fixture 做 diff。归一化器已支持复数会话 id 并会折叠任何游离 UUID，因此无需修改归一化器。
+`harvestSessionLogs` 递归收集 sessions 根目录下所有固定命名为 `session.jsonl` 的 transcript（JSONL 后端为每个父会话和子会话分别提供独立的项目/会话目录），解析各自的 header，并按主会话优先排序：顶层会话（无 `parentSession`）在前，各子会话按 `createdAt` 升序排列。`RunResult.sessionLogs` 是复数结果；spec 在录制时将每份日志写回对应 fixture（`session.jsonl` + `session.<n>.jsonl`），在回放时将每份收集到的日志与其 fixture 做 diff。归一化器已支持复数会话 id 并会折叠任何游离 UUID，因此无需修改归一化器。
 
 ### 4. 场景