Merge remote-tracking branch 'refs/remotes/origin/master' into worktree/retarget-pr961-20260808

# Conflicts: # .agents/notes/implemented/bug-fix/2026-07-27-stable-snapshot-refresh-volatiles.i18n.yaml # .agents/notes/implemented/bug-fix/2026-07-27-stable-snapshot-refresh-volatiles.zh.md # apps/web/tests/scaffold.ts # examples/tui-agent/tests/tui.snapshot.ts # packages/support/acp-snapshot/README.i18n.yaml # packages/support/acp-snapshot/README.md # packages/support/acp-snapshot/README.zh.md
2026-08-08 01:49:04 +08:00
parent 913d9cf9d8 a29a0a4773
commit a29f55881e
4422 changed files with 272205 additions and 60192 deletions
@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write packages/support/acp-snapshot/README.md
-README.md: c5a1a07b9f85e1a91c52fe102be17e4172be112d
-README.zh.md: 2454df9b3d8e67b4728d6582279ba21798d2ba9e
+README.md: 9d142dc964e60f9508b6c137525eb916cdbd969f
+README.zh.md: 2e88e7c5bb8acb0cd99a35b5fe0fbfc15d6101e3
@@ -8,8 +8,8 @@ Four layers, importable separately:

 - **`launchAcpTestAgent` (launcher)** — boots a source agent under tsx or a built `lib` agent under plain Node from a supplied cwd, connects the SDK client over a raw-byte stdout tee, collects session updates and stderr, surfaces asynchronous spawn failures through startup, fails closed on unhandled permission requests, and owns graceful or signalled shutdown. Shutdown waits for process exit, inherited stdio closure, and ACP parser exhaustion before resolving or propagating a child error, so captures are complete and callers can remove owned paths after either outcome. When Windows accepts forced termination but publishes its exit marker asynchronously, shutdown gives that marker a bounded grace before treating fallback refusal as a second failure. Snapshot and ordinary e2e suites share this process boundary; a test supplies only agent paths, cwd, environment overrides, and any permission policy.
 - **`runScenario` (harness)** — drives ACP JSON-RPC stdio from a deterministic `input.json` script through the launcher, tees raw stdout for the expected-output and purity checks, and harvests every persisted raw JSONL session log (parent and subagent children, primary-first) after graceful stdin EOF. `AgentUnderTest` supplies absolute `binScript`, optional `libBinScript`, `configPath`, and `tsconfigPath` paths because the subprocess cwd is outside the repo; `workspaceParent` may move the generated child cwd from the platform temp directory when that grant is itself under test. Startup failures preserve captured agent stderr in the rejected diagnostic.
- **Normalizers** — pure functions turning captured surfaces into stable text or portable fixtures: `normalizeStdout` (JSON-RPC ids → first-seen sequence; UUIDs and every native/JavaScript filesystem spelling of the generated cwd → tokens, longest-first; cwd-rooted separators selected as canonical `/` or host-native; doubles as the stdout-purity check), `normalizeSessionLog` (times zeroed, `seq` kept, the same cwd-path policy), `tokenizeSessionFixtureCwd` (the generated workspace and its filesystem aliases → `{{cwd}}`, authored temp paths unchanged), `scrubSystemPrompts` (prompt text → `{{system}}`), `scrubToolSchemas` (schema bulk → `{{tools}}`), `scrubRequestHeaders` (all header bulk → `{{system}}`/`{{tools}}`/`{{messagePrefix}}` outside each pin, structure kept — [pinned-header Agent Note](../../../.agents/notes/archived/testing/2026-07-06-pin-request-header-content-in-one-scenario.md)), and `stabilizeFixtureMessageIds` (committed UUIDs carried into unchanged, unambiguous messages across any recorder's fixture-ready parent/child logs).
- **`defineAcpSnapshotSuite` (factory)** — registers the whole describe/it tree for a scenario table: per-scenario expected-output and re-persisted-log comparisons, record/refresh fixture write-back, rejection of structured `UNKNOWN_TOOL` results, a tokenized pin per header class composed with independently shared `system-prompt.expected.md` and `tool-schemas.expected.json` sidecars, and a live uniformity guard. Its fixture guards reject orphan scenario dirs, missing files, multiple pins for one class, duplicate sidecar content, unscrubbed JSONL headers, and malformed pinning headers. Before record or refresh writes fixtures, an unchanged complete message retains its committed UUID when its identity-free value resolves to exactly one fresh ID and one existing ID across the scenario's parent/child logs; new, changed, and ambiguous messages keep fresh UUIDs. Refresh evaluates fresh leaves with the harvested run's ids, cwd, and every cwd alias, then reuses normalized-equivalent leaves only when the complete logical-record layout aligns and volatile string replacements form a bijection; ambiguous logs keep fresh strings, and fresh semantic values remain authoritative. It also expands packed timing envelopes before aligning event times, so switching between packed and unpacked layouts cannot shift later records. A newly inserted `session/title` receives its preceding event's time so feature-driven insertions do not churn the remainder of a fixture. Each scenario directory's `session.jsonl` plus contiguous `session.<n>.jsonl` siblings are the ordered primary/child inventory; the scenario table does not duplicate their count. Must be called at vitest collection time.
+- **Normalizers** — pure functions turning captured surfaces into stable text or portable fixtures: `normalizeStdout` (JSON-RPC ids → first-seen sequence; UUIDs and every native/JavaScript filesystem spelling of the generated cwd → tokens, longest-first; cwd-rooted separators selected as canonical `/` or host-native; doubles as the stdout-purity check), `normalizeSessionLog` (times zeroed, `seq` kept, the same cwd-path policy), `tokenizeSessionFixtureCwd` (the generated workspace and its filesystem aliases → one canonical `{{cwd}}`, including an already-tokenized macOS `/private` alias; authored temp paths unchanged), `scrubSystemPrompts` (prompt text → `{{system}}`), `scrubToolSchemas` (schema bulk → `{{tools}}`), `scrubRequestHeaders` (all header bulk → `{{system}}`/`{{tools}}`/`{{messagePrefix}}` outside each pin, structure kept — [pinned-header Agent Note](../../../.agents/notes/archived/testing/2026-07-06-pin-request-header-content-in-one-scenario.md)), and `stabilizeFixtureMessageIds` (committed UUIDs carried into unchanged, unambiguous messages across any recorder's fixture-ready parent/child logs).
+- **`defineAcpSnapshotSuite` (factory)** — registers the whole describe/it tree for a scenario table: per-scenario expected-output and re-persisted-log comparisons, record/refresh fixture write-back, rejection of structured `UNKNOWN_TOOL` results, a tokenized pin per header class composed with independently shared `system-prompt.expected.md` and `tool-schemas.expected.json` sidecars, and a live uniformity guard. Its fixture guards reject orphan scenario dirs, missing files, multiple pins for one class, duplicate sidecar content, noncanonical macOS-prefixed cwd tokens, unscrubbed JSONL headers, and malformed pinning headers. Before record or refresh writes fixtures, an unchanged complete message retains its committed UUID when its identity-free value resolves to exactly one fresh ID and one existing ID across the scenario's parent/child logs; new, changed, and ambiguous messages keep fresh UUIDs. Refresh evaluates fresh leaves with the harvested run's ids, cwd, and every cwd alias, then reuses normalized-equivalent leaves only when the complete logical-record layout aligns and volatile string replacements form a bijection; ambiguous logs keep fresh strings, and fresh semantic values remain authoritative. It also expands packed timing envelopes before aligning event times, so switching between packed and unpacked layouts cannot shift later records. A newly inserted `session/title` receives its preceding event's time so feature-driven insertions do not churn the remainder of a fixture. Each scenario directory's `session.jsonl` plus contiguous `session.<n>.jsonl` siblings are the ordered primary/child inventory; the scenario table does not duplicate their count. Must be called at vitest collection time.

 Committed session fixtures use canonical packed rows. An in-flight branch that merges this contract runs the [temporary repository migrator](../../../scripts/migrate-packed-session-fixtures.ts) with `pnpm run migrate:packed-session-fixtures`; its [removal proposal](../../../.agents/notes/proposed/process/2026-07-26-remove-packed-session-fixture-migrator.md) owns deletion after affected branches converge.

@@ -55,11 +55,11 @@ A scenario booting a differently-composed tree sets its own `configPath` (an ove

 A pin owns its generated `system-prompt.expected.md` or `tool-schemas.expected.json` by default; `systemPromptSource` and `toolSchemasSource` name another pin when the complete corresponding sequence is identical, so each distinct version is committed once. The pin's `session.jsonl` stores `"system":"{{system}}","tools":"{{tools}}"` while retaining config, reason, and any model-visible prefix. A pin with legitimate mid-run header changes declares `expectedHeaderChanges`; a shared source must declare the same count, and record/refresh rejects claimants that generate different bytes.

-Every scenario compares `stdout.expected.jsonl` with cwd-rooted separators canonicalized to `/`. On Windows, `pinsNativeWindowsStdout` additionally compares the complete `stdout.expected.windows.jsonl` after the shared expected output and requires that sidecar exactly when enabled. A scenario requiring a non-Windows host declares `posixOnly`, which skips its run test on Windows while the fixture guards keep covering its committed files everywhere; examples include POSIX process semantics (e.g. cancelling a live bash call kills a detached process group) and generated paths Windows cannot represent.
+Every scenario compares `stdout.expected.jsonl` with cwd-rooted separators canonicalized to `/`. On Windows, `pinsNativeWindowsStdout` additionally compares the complete `stdout.expected.windows.jsonl` after the shared expected output and requires that sidecar exactly when enabled. A scenario requiring a non-Windows host declares `posixOnly`, which skips its run test on Windows while the fixture guards keep covering its committed files everywhere; examples include POSIX process semantics (e.g. cancelling a live bash call kills a detached process group) and generated paths Windows cannot represent. A scenario whose composition needs a usable `pwsh` declares `pwshOnly`; the caller-supplied `hasPwsh` probe (the shipped acp-agent suite follows the executor's own resolution, so Program Files installs count) skips the run test when no usable `pwsh` resolves while the fixture guards keep covering its committed files everywhere.

 The example also ships a `cordis.snapshot.yml` replay overlay next to its `cordis.yml` (the bin swaps them under `DSH_SNAPSHOT=replay` — [single-source replay config Agent Note](../../../.agents/notes/archived/testing/2026-07-04-single-source-acp-replay-config.md)); replay fixtures are served by [`dsh-llm-replay`](../llm-replay/README.md), which this package points at via the `DSH_SNAPSHOT_*` env vars it sets on the child. `pnpm run test:snapshot:record` calls the live LLM and rewrites the recorded scenarios' model fixtures; `pnpm run test:snapshot:refresh` stays keyless, runs the replay overlay, and rewrites stdout, comparable session-log expected outputs, and owned prompt and tool-schema sidecars from the committed model scripts. Fixture roles, record/replay/refresh semantics, and scenario-table fields are documented on `Scenario` and in the [snapshot Agent Note](../../../.agents/notes/implemented/testing/2026-06-19-acp-snapshot-tests.md).

-Constraints: `suite.ts` and `harness.ts` import vitest (the harness polls its durable-boundary waits through `vi.waitFor`), so the package entry is importable only inside a vitest run (the launcher and normalizers have no such dependency but ship from the same entry). The launcher and suite factory are ACP-specific by design — the launcher speaks the SDK's `ClientSideConnection` — while the normalizers are transport-neutral session-log/text helpers also consumed by the JSON-RPC, TUI, and web snapshot recorders. Input scripts cover initialization, fresh-session creation, text prompting, cancellation, expected RPC failures, and durable turn-boundary waits. Permission round-trips are a FIFO queue of option-kind selections (`allow_once`, `reject_once`, …) mapped to the agent-issued `optionId`; an absent or exhausted queue answers `cancelled`, and an unoffered kind rejects the run.
+Constraints: `suite.ts` and `harness.ts` import vitest (the harness polls its durable-boundary waits through `vi.waitFor`), so the package entry is importable only inside a vitest run (the launcher and normalizers have no such dependency but ship from the same entry). The launcher and suite factory are ACP-specific by design — the launcher speaks the SDK's `ClientSideConnection` — while the normalizers are transport-neutral session-log/text helpers also consumed by the JSON-RPC and Web snapshot recorders. Input scripts cover initialization, fresh-session creation, text prompting, cancellation, expected RPC failures, and durable turn-boundary waits. Permission round-trips are a FIFO queue of option-kind selections (`allow_once`, `reject_once`, …) mapped to the agent-issued `optionId`; an absent or exhausted queue answers `cancelled`, and an unoffered kind rejects the run.

 ## Model Experience

@@ -8,8 +8,8 @@ ACP 快照套件工具包：无密钥快照层（`pnpm run test:snapshot`，见[

 - **`launchAcpTestAgent`（启动器）**：从指定 cwd 在 tsx 下启动源 agent，或在普通 Node 下启动已构建 `lib` agent；通过原始字节 stdout tee 连接 SDK 客户端，收集会话更新和 stderr，在启动过程中公开异步 spawn 失败，对未处理权限请求快速失败，并负责优雅或带信号关闭。关闭会等待进程退出、继承 stdio 关闭和 ACP parser 耗尽，然后才解析或传播子级错误，使捕获内容完整，且调用方可在任一结果后移除自有路径。当 Windows 接受强制终止但异步发布退出标记时，关闭会给该标记有界宽限，然后才将回退拒绝视为第二次失败。快照和普通 e2e 套件共享该进程边界；测试只需提供 agent 路径、cwd、环境覆盖和任何权限策略。
 - **`runScenario`（harness）**：通过启动器从确定性 `input.json` 脚本驱动 ACP JSON-RPC stdio，将原始 stdout tee 给预期输出和纯度检查，并在优雅 stdin EOF 后收集每个持久化原始 JSONL 会话日志（父级和 subagent 子级，主级优先）。`AgentUnderTest` 提供绝对 `binScript`、可选 `libBinScript`、`configPath` 和 `tsconfigPath` 路径，因为子进程 cwd 位于仓库外。当生成子级 cwd 自身位于待测授权中时，`workspaceParent` 可以将它从平台临时目录移出。启动失败会在拒绝诊断中保留已捕获 agent stderr。
- **规范化器**：将已捕获接口转换为稳定文本或可移植 fixture 的纯函数：`normalizeStdout`（JSON-RPC id → 首次出现序列；UUID 以及生成 cwd 的每个原生/JavaScript 文件系统写法 → token，按最长优先；根据 cwd 的分隔符选择规范 `/` 或宿主原生形式；同时作为 stdout 纯度检查）、`normalizeSessionLog`（时间归零、保留 `seq`、使用同一 cwd 路径策略）、`tokenizeSessionFixtureCwd`（生成的 workspace 及其文件系统别名 → `{{cwd}}`，手工编写的临时路径保持不变）、`scrubSystemPrompts`（提示词文本 → `{{system}}`）、`scrubToolSchemas`（schema bulk → `{{tools}}`）、`scrubRequestHeaders`（每个 pin 之外的所有 header bulk → `{{system}}`/`{{tools}}`/`{{messagePrefix}}`，保留结构；见[header 固定 Agent Note](../../../.agents/notes/archived/testing/2026-07-06-pin-request-header-content-in-one-scenario.md)）和 `stabilizeFixtureMessageIds`（针对任意录制器已准备写入 fixture 的父级/子级日志，将已提交 UUID 带入未变化且无歧义的消息）。
- **`defineAcpSnapshotSuite`（工厂）**：为场景表注册完整 describe/it 树：每场景预期输出与重新持久化日志比较、录制/刷新 fixture 回写、拒绝结构化 `UNKNOWN_TOOL` 结果、每个 header 类别一个 token 化 pin（由可独立共享的 `system-prompt.expected.md` 和 `tool-schemas.expected.json` sidecar 组合而成），以及实时一致性保护。其 fixture 保护会拒绝遗留场景目录、缺失文件、一个类别包含多个 pin、重复的 sidecar 内容、未擦除的 JSONL header，以及格式错误的 pin header。在录制或刷新写入 fixture 前，如果一条未变化的完整消息去除身份后的值在场景的父级/子级日志中恰好对应一个本次生成的 ID 和一个现有 ID，它就会保留已提交的 UUID；新增、发生变化和有歧义的消息则保留本次生成的 UUID。刷新会使用收集所得本次运行的 id、cwd 及全部 cwd 别名评估本次生成的叶值；只有完整逻辑记录布局对齐且易变字符串替换形成双射时，才会复用规范化后等价的叶值；有歧义的日志保留本次生成的字符串，而本次生成的语义值仍为权威数据。它还会在对齐事件时间前展开打包时序 envelope，因此切换打包/非打包布局无法移动后续记录。新插入的 `session/title` 使用前一个事件的时间，因此功能驱动的插入不会扰动 fixture 余下部分。每个场景目录的 `session.jsonl` 和连续 `session.<n>.jsonl` 同级文件是有序主级/子级清单；场景表不重复其数量。必须在 vitest 收集时调用。
+- **规范化器**：将已捕获接口转换为稳定文本或可移植 fixture 的纯函数：`normalizeStdout`（JSON-RPC id → 首次出现序列；UUID 以及生成 cwd 的每个原生/JavaScript 文件系统写法 → token，按最长优先；根据 cwd 的分隔符选择规范 `/` 或宿主原生形式；同时作为 stdout 纯度检查）、`normalizeSessionLog`（时间归零、保留 `seq`、使用同一 cwd 路径策略）、`tokenizeSessionFixtureCwd`（生成的 workspace 及其文件系统别名，包括已 token 化的 macOS `/private` 别名 → 单一规范 `{{cwd}}`；手工编写的临时路径保持不变）、`scrubSystemPrompts`（提示词文本 → `{{system}}`）、`scrubToolSchemas`（schema bulk → `{{tools}}`）、`scrubRequestHeaders`（每个 pin 之外的所有 header bulk → `{{system}}`/`{{tools}}`/`{{messagePrefix}}`，保留结构；见[header 固定 Agent Note](../../../.agents/notes/archived/testing/2026-07-06-pin-request-header-content-in-one-scenario.md)）和 `stabilizeFixtureMessageIds`（针对任意录制器已准备写入 fixture 的父级/子级日志，将已提交 UUID 带入未变化且无歧义的消息）。
+- **`defineAcpSnapshotSuite`（工厂）**：为场景表注册完整 describe/it 树：每场景预期输出与重新持久化日志比较、录制/刷新 fixture 回写、拒绝结构化 `UNKNOWN_TOOL` 结果、每个 header 类别一个 token 化 pin（由可独立共享的 `system-prompt.expected.md` 和 `tool-schemas.expected.json` sidecar 组合而成），以及实时一致性保护。其 fixture 保护会拒绝遗留场景目录、缺失文件、一个类别包含多个 pin、重复的 sidecar 内容、带非规范 macOS 前缀的 cwd token、未擦除的 JSONL header，以及格式错误的 pin header。在录制或刷新写入 fixture 前，如果一条未变化的完整消息去除身份后的值在场景的父级/子级日志中恰好对应一个本次生成的 ID 和一个现有 ID，它就会保留已提交的 UUID；新增、发生变化和有歧义的消息则保留本次生成的 UUID。刷新会使用收集所得本次运行的 id、cwd 及全部 cwd 别名评估本次生成的叶值；只有完整逻辑记录布局对齐且易变字符串替换形成双射时，才会复用规范化后等价的叶值；有歧义的日志保留本次生成的字符串，而本次生成的语义值仍为权威数据。它还会在对齐事件时间前展开打包时序 envelope，因此切换打包/非打包布局无法移动后续记录。新插入的 `session/title` 使用前一个事件的时间，因此功能驱动的插入不会扰动 fixture 余下部分。每个场景目录的 `session.jsonl` 和连续 `session.<n>.jsonl` 同级文件是有序主级/子级清单；场景表不重复其数量。必须在 vitest 收集时调用。

 签入仓库的会话 fixture 使用规范打包行。合并此契约的在途分支通过 `pnpm run migrate:packed-session-fixtures` 运行[临时仓库迁移器](../../../scripts/migrate-packed-session-fixtures.ts)；待受影响分支收敛后，由其[移除提案](../../../.agents/notes/proposed/process/2026-07-26-remove-packed-session-fixture-migrator.md)负责删除该迁移器。

@@ -55,11 +55,11 @@ defineAcpSnapshotSuite({

 每个 pin 默认拥有其生成的 `system-prompt.expected.md` 或 `tool-schemas.expected.json`；当完整的对应序列相同时，`systemPromptSource` 和 `toolSchemasSource` 指定另一个 pin 作为来源，因此每个不同版本只提交一次。该 pin 的 `session.jsonl` 存储 `"system":"{{system}}","tools":"{{tools}}"`，同时保留配置、原因和任何模型可见前缀。具有合法运行中 header 变更的 pin 声明 `expectedHeaderChanges`；共享来源必须声明相同的 header 变更数量，录制/刷新会拒绝生成不同字节的共享引用方。

-每个场景都比较 `stdout.expected.jsonl`，其中以 cwd 为根的分隔符规范化为 `/`。在 Windows 上，`pinsNativeWindowsStdout` 还会在共享预期输出之后比较完整 `stdout.expected.windows.jsonl`，并在启用时精确要求该 sidecar。需要非 Windows 主机的场景声明 `posixOnly`，在 Windows 上跳过运行测试，但 fixture 保护仍在所有平台覆盖其已提交文件；示例包括 POSIX 进程语义（例如取消实时 bash 调用会终止脱离进程组）和 Windows 无法表示的生成路径。
+每个场景都比较 `stdout.expected.jsonl`，其中以 cwd 为根的分隔符规范化为 `/`。在 Windows 上，`pinsNativeWindowsStdout` 还会在共享预期输出之后比较完整 `stdout.expected.windows.jsonl`，并在启用时精确要求该 sidecar。需要非 Windows 主机的场景声明 `posixOnly`，在 Windows 上跳过运行测试，但 fixture 保护仍在所有平台覆盖其已提交文件；示例包括 POSIX 进程语义（例如取消实时 bash 调用会终止脱离进程组）和 Windows 无法表示的生成路径。组合需要可用 `pwsh` 的场景声明 `pwshOnly`；调用方提供的 `hasPwsh` 探测（随附的 acp-agent 套件遵循执行器自身的解析，因此 Program Files 安装也计入）在解析不到可用 `pwsh` 时跳过运行测试，而 fixture 保护仍处处覆盖其已提交文件。

 示例还发布 `cordis.snapshot.yml` 回放 overlay，位于 `cordis.yml` 旁边（bin 在 `DSH_SNAPSHOT=replay` 下交换它们，见[单源回放配置 Agent Note](../../../.agents/notes/archived/testing/2026-07-04-single-source-acp-replay-config.md)）；回放 fixture 由 [`dsh-llm-replay`](../llm-replay/README.md) 提供，该包通过对子级设置的 `DSH_SNAPSHOT_*` env var 指向它。`pnpm run test:snapshot:record` 调用实时 LLM，并重写已记录场景的模型 fixture；`pnpm run test:snapshot:refresh` 保持无密钥，运行回放 overlay，并从已提交模型脚本重写 stdout、可比较会话日志预期输出，以及各 pin 自有的提示词与工具 schema sidecar。Fixture 角色、录制/回放/刷新语义和场景表字段记录在 `Scenario` 以及[快照 Agent Note](../../../.agents/notes/implemented/testing/2026-06-19-acp-snapshot-tests.md) 中。

-约束：`suite.ts` 与 `harness.ts` 导入 vitest（harness 通过 `vi.waitFor` 轮询其持久边界等待），因此包入口只能在 vitest 运行中导入（启动器和规范化器没有此依赖，但从同一入口发布）。启动器和套件工厂按设计专用于 ACP，启动器使用 SDK 的 `ClientSideConnection`；规范化器是与传输无关的会话日志/文本辅助工具，还由 JSON-RPC、TUI 和 web 快照录制器消费。输入脚本覆盖初始化、新建会话、文本提示、取消、预期 RPC 失败和持久轮次边界等待。权限往返是选项类别选择（`allow_once`、`reject_once`等）的 FIFO 队列，映射到 agent 发出的 `optionId`；缺少或耗尽的队列回答 `cancelled`，未提供类别会拒绝运行。
+约束：`suite.ts` 与 `harness.ts` 导入 vitest（harness 通过 `vi.waitFor` 轮询其持久边界等待），因此包入口只能在 vitest 运行中导入（启动器和规范化器没有此依赖，但从同一入口发布）。启动器和套件工厂按设计专用于 ACP，启动器使用 SDK 的 `ClientSideConnection`；规范化器是与传输无关的会话日志/文本辅助工具，还由 JSON-RPC 和 Web 快照录制器消费。输入脚本覆盖初始化、新建会话、文本提示、取消、预期 RPC 失败和持久轮次边界等待。权限往返是选项类别选择（`allow_once`、`reject_once` 等）的 FIFO 队列，映射到 agent 发出的 `optionId`；缺少或耗尽的队列回答 `cancelled`，未提供类别会拒绝运行。

 ## 模型体验

@@ -21,9 +21,7 @@
  "files": [
    "lib/index.js",
    "lib/invariant.js",
-    "lib/types/**/*.d.ts",
-    "lib/types/**/*.d.ts.map",
-    "src"
+    "lib/types/**/*.d.ts"
  ],
  "license": "BSD-3-Clause",
  "dependencies": {
@@ -43,14 +43,24 @@ const WAIT_POLL_INTERVAL_MS = 10
 * reference, since a committed file cannot know the id in advance.
 *
 * `promptAndCancel` starts a prompt without awaiting completion, waits for a
- * readiness condition, then cancels and awaits completion. `waitForFile`
- * observes a cwd-relative marker; the default observes the durable turn start.
+ * readiness condition, then cancels and awaits completion. Its optional
+ * `waitForFile` observes a cwd-relative marker; otherwise it waits for the
+ * durable turn start. The standalone `waitForFile` holds the next script step
+ * behind the same marker.
 * `promptAndWaitForAgentMessage` arms an exact text-chunk waiter before sending
 * the prompt, then keeps the application live until that later update arrives.
 * `waitForTurnStart` waits for an open durable turn, optionally at or beyond a
 * specified turn number. `waitForTurnEnd` holds the subprocess open until the
 * selected session's latest complete raw-JSONL turn boundary is `turn/end`.
+ * `waitForGoalPhase` waits for the latest durable goal snapshot to reach one phase.
+ * `waitForInboxMessage` waits for inserted inbox text containing a scenario marker.
+ * `waitForSubagentTurnEnd` waits until one background child has persisted a
+ * closed model-work turn after its own descriptor; child progress has no ACP
+ * update to wait on.
 * `waitForTitleAfterTurnEnd` additionally waits for a later durable title.
+ * `waitForEventAfterTurnEnd` waits until a complete record of the given event
+ * type follows the latest closed turn — for scenarios whose asserted state
+ * (e.g. a goal pause) is appended only after cancellation reaches idle.
 * A standalone `cancel` may also wait for a cwd-relative readiness marker.
 * All wait timeouts default to 10s.
 */
@@ -66,9 +76,14 @@ export type InputStep =
    text: string
    waitForFile?: { path: string; timeoutMs?: number }
  }
+  | { op: 'waitForFile'; path: string; timeoutMs?: number }
  | { op: 'waitForTurnStart'; minimumTurn?: number; timeoutMs?: number }
  | { op: 'waitForTurnEnd'; timeoutMs?: number }
+  | { op: 'waitForSubagentTurnEnd'; child?: number; minimumTurn?: number; timeoutMs?: number }
+  | { op: 'waitForGoalPhase'; phase: 'active' | 'paused' | 'blocked' | 'complete'; timeoutMs?: number }
+  | { op: 'waitForInboxMessage'; text: string; timeoutMs?: number }
  | { op: 'waitForTitleAfterTurnEnd'; timeoutMs?: number }
+  | { op: 'waitForEventAfterTurnEnd'; type: string; timeoutMs?: number }
  | { op: 'cancel'; waitForFile?: { path: string; timeoutMs?: number } }

 /** A scenario's `input.json`: an ordered list of input steps. */
@@ -290,7 +305,11 @@ export async function runScenario(input: InputScript, opts: RunOptions): Promise
        (id) => { sessionId = id },
        (id, timeoutMs, minimumTurn) => waitForPersistedTurnStart(sessionsRoot, id, timeoutMs, minimumTurn),
        (id, timeoutMs) => waitForPersistedTurnEnd(sessionsRoot, id, timeoutMs),
+        (child, timeoutMs, minimumTurn) => waitForPersistedChildTurnEnd(sessionsRoot, child, timeoutMs, minimumTurn),
+        (id, phase, timeoutMs) => waitForPersistedGoalPhase(sessionsRoot, id, phase, timeoutMs),
+        (id, text, timeoutMs) => waitForPersistedInboxMessage(sessionsRoot, id, text, timeoutMs),
        (id, timeoutMs) => waitForPersistedTitleAfterTurnEnd(sessionsRoot, id, timeoutMs),
+        (id, type, timeoutMs) => waitForPersistedEventAfterTurnEnd(sessionsRoot, id, type, timeoutMs),
      )
      // A permission exchange happens while a step's request is in flight, so
      // by the time the step settles any script bug it exposed is captured —
@@ -364,7 +383,11 @@ async function runStep(
  setSessionId: (id: string) => void,
  waitForTurnStart: (sessionId: string, timeoutMs?: number, minimumTurn?: number) => Promise<void>,
  waitForTurnEnd: (sessionId: string, timeoutMs?: number) => Promise<void>,
+  waitForChildTurnEnd: (child: number, timeoutMs?: number, minimumTurn?: number) => Promise<void>,
+  waitForGoalPhase: (sessionId: string, phase: string, timeoutMs?: number) => Promise<void>,
+  waitForInboxMessage: (sessionId: string, text: string, timeoutMs?: number) => Promise<void>,
  waitForTitleAfterTurnEnd: (sessionId: string, timeoutMs?: number) => Promise<void>,
+  waitForEventAfterTurnEnd: (sessionId: string, type: string, timeoutMs?: number) => Promise<void>,
 ): Promise<void> {
  switch (step.op) {
    case 'initialize':
@@ -436,18 +459,42 @@ async function runStep(
      await promptDone
      return
    }
+    case 'waitForFile':
+      await waitForWorkspaceFile(cwd, step.path, step.timeoutMs)
+      return
    case 'waitForTurnEnd': {
      const sessionId = getSessionId()
      if (sessionId === undefined) throw new Error('snapshot-harness: waitForTurnEnd before newSession')
      await waitForTurnEnd(sessionId, step.timeoutMs)
      return
    }
+    case 'waitForSubagentTurnEnd':
+      await waitForChildTurnEnd(step.child ?? 1, step.timeoutMs, step.minimumTurn)
+      return
+    case 'waitForGoalPhase': {
+      const sessionId = getSessionId()
+      if (sessionId === undefined) throw new Error('snapshot-harness: waitForGoalPhase before newSession')
+      await waitForGoalPhase(sessionId, step.phase, step.timeoutMs)
+      return
+    }
+    case 'waitForInboxMessage': {
+      const sessionId = getSessionId()
+      if (sessionId === undefined) throw new Error('snapshot-harness: waitForInboxMessage before newSession')
+      await waitForInboxMessage(sessionId, step.text, step.timeoutMs)
+      return
+    }
    case 'waitForTitleAfterTurnEnd': {
      const sessionId = getSessionId()
      if (sessionId === undefined) throw new Error('snapshot-harness: waitForTitleAfterTurnEnd before newSession')
      await waitForTitleAfterTurnEnd(sessionId, step.timeoutMs)
      return
    }
+    case 'waitForEventAfterTurnEnd': {
+      const sessionId = getSessionId()
+      if (sessionId === undefined) throw new Error('snapshot-harness: waitForEventAfterTurnEnd before newSession')
+      await waitForEventAfterTurnEnd(sessionId, step.type, step.timeoutMs)
+      return
+    }
    case 'waitForTurnStart': {
      const sessionId = getSessionId()
      if (sessionId === undefined) throw new Error('snapshot-harness: waitForTurnStart before newSession')
@@ -515,6 +562,95 @@ async function waitForPersistedTurnEnd(
  }, { interval: WAIT_POLL_INTERVAL_MS, timeout: timeoutMs })
 }

+/**
+ * Wait until the Nth harvested child Session closes a model work turn.
+ *
+ * Harvest order matches `session.1.jsonl`, `session.2.jsonl`, and so on. A
+ * continuable child appends its descriptor after any inherited history and
+ * before accepting its first prompt, so only a later request header proves its
+ * own model work reached a closed turn.
+ */
+async function waitForPersistedChildTurnEnd(
+  root: string,
+  child: number,
+  timeoutMs = DEFAULT_WAIT_TIMEOUT_MS,
+  minimumTurn = 1,
+): Promise<void> {
+  await vi.waitFor(async () => {
+    const log = (await harvestSessionLogs(root))[child]
+    if (log === undefined || !latestTurnIsClosed(log.content)
+      || !hasRequestHeaderAfterDescriptor(log.content)
+      || !hasClosedTurn(log.content, minimumTurn)) {
+      throw new Error(
+        `snapshot-harness: subagent child #${child} did not persist closed turn ${minimumTurn} within ${timeoutMs}ms`,
+      )
+    }
+  }, { interval: WAIT_POLL_INTERVAL_MS, timeout: timeoutMs })
+}
+
+/** Whether a raw session log contains the requested closed turn. */
+function hasClosedTurn(content: string, turn: number): boolean {
+  return content.split('\n').filter(Boolean).some((line) => {
+    const event = JSON.parse(line) as { type?: unknown; data?: { turn?: unknown } }
+    return event.type === 'turn/end' && event.data?.turn === turn
+  })
+}
+
+/** Wait until the latest durable goal snapshot reaches one phase. */
+async function waitForPersistedGoalPhase(
+  root: string,
+  sessionId: string,
+  phase: string,
+  timeoutMs = DEFAULT_WAIT_TIMEOUT_MS,
+): Promise<void> {
+  await vi.waitFor(async () => {
+    const content = (await harvestSessionLogs(root)).find(log => log.id === sessionId)?.content
+    const matched = content?.split('\n').filter(Boolean).some((line) => {
+      const event = JSON.parse(line) as { type?: unknown; data?: { goal?: { phase?: unknown } } }
+      return event.type === 'goal/change' && event.data?.goal?.phase === phase
+    }) ?? false
+    if (!matched) {
+      throw new Error(`snapshot-harness: session "${sessionId}" did not persist goal phase "${phase}" within ${timeoutMs}ms`)
+    }
+  }, { interval: WAIT_POLL_INTERVAL_MS, timeout: timeoutMs })
+}
+
+/** Wait until an inserted inbox message contains scenario-owned text. */
+async function waitForPersistedInboxMessage(
+  root: string,
+  sessionId: string,
+  text: string,
+  timeoutMs = DEFAULT_WAIT_TIMEOUT_MS,
+): Promise<void> {
+  await vi.waitFor(async () => {
+    const log = (await harvestSessionLogs(root)).find(candidate => candidate.id === sessionId)
+    const matched = log?.content.split('\n').some((line) => {
+      if (line.length === 0) return false
+      const record = JSON.parse(line) as {
+        type?: unknown
+        data?: { inserted?: Array<{ content?: Array<{ type?: unknown; text?: unknown }> }> }
+      }
+      return record.type === 'agent/inbox/spliced' && record.data?.inserted?.some(message =>
+        message.content?.some(block => block.type === 'text'
+          && typeof block.text === 'string' && block.text.includes(text))) === true
+    }) ?? false
+    if (!matched) {
+      throw new Error(`snapshot-harness: session "${sessionId}" did not persist expected inbox message within ${timeoutMs}ms`)
+    }
+  }, { interval: WAIT_POLL_INTERVAL_MS, timeout: timeoutMs })
+}
+
+/** Whether a child log contains model work after its own descriptor event. */
+function hasRequestHeaderAfterDescriptor(content: string): boolean {
+  const events = content.slice(0, content.lastIndexOf('\n') + 1)
+    .split('\n')
+    .filter(line => line.length > 0)
+    .map(line => JSON.parse(line) as { type?: unknown })
+  const descriptor = events.findLastIndex(event => event.type === 'subagent/descriptor')
+  return descriptor >= 0
+    && events.slice(descriptor + 1).some(event => event.type === 'request/header')
+}
+
 /** Wait until a complete provider or fallback title record follows the latest closed turn. */
 async function waitForPersistedTitleAfterTurnEnd(
  root: string,
@@ -529,6 +665,21 @@ async function waitForPersistedTitleAfterTurnEnd(
  }, { interval: WAIT_POLL_INTERVAL_MS, timeout: timeoutMs })
 }

+/** Wait until a complete record of `type` follows the latest closed turn. */
+async function waitForPersistedEventAfterTurnEnd(
+  root: string,
+  sessionId: string,
+  type: string,
+  timeoutMs = DEFAULT_WAIT_TIMEOUT_MS,
+): Promise<void> {
+  await vi.waitFor(async () => {
+    const log = (await harvestSessionLogs(root)).find(candidate => candidate.id === sessionId)
+    if (log === undefined || !latestEventFollowsTurnEnd(log.content, type)) {
+      throw new Error(`snapshot-harness: session "${sessionId}" did not persist ${type} after turn/end within ${timeoutMs}ms`)
+    }
+  }, { interval: WAIT_POLL_INTERVAL_MS, timeout: timeoutMs })
+}
+
 /** Wait for a cwd-relative marker proving an external action reached readiness. */
 async function waitForWorkspaceFile(
  cwd: string,
@@ -557,6 +708,13 @@ function latestTitleFollowsTurnEnd(content: string): boolean {
  return turnEnd >= 0 && complete.lastIndexOf('\n{"type":"session/title",') > turnEnd
 }

+/** Return whether a complete record of `type` occurs after the last complete turn end. */
+function latestEventFollowsTurnEnd(content: string, type: string): boolean {
+  const complete = content.slice(0, content.lastIndexOf('\n') + 1)
+  const turnEnd = complete.lastIndexOf('\n{"type":"turn/end",')
+  return turnEnd >= 0 && complete.lastIndexOf(`\n{"type":"${type}",`) > turnEnd
+}
+
 /** Return the latest open turn number, validating the persisted boundary record. */
 function latestOpenTurn(content: string): number | undefined {
  const complete = content.slice(0, content.lastIndexOf('\n') + 1)
@@ -197,7 +197,7 @@ function tokenizeFixtureString(value: string, ctx: NormalizeContext, basename: s
    + String.raw`(?=$|[\\/\s<>'"()\[\]{},;:!?=])`,
    'g',
  )
-  return exact.replace(absoluteCwd, CWD)
+  return exact.replace(absoluteCwd, CWD).split(`/private${CWD}`).join(CWD)
 }

 /** Recursively replace generated-cwd spellings while preserving every other JSON value. */
@@ -40,6 +40,11 @@ const SYSTEM_PROMPT_SNAPSHOT = 'system-prompt.expected.md'
 /** The structured tool-schema snapshot beside its owning header pin. */
 const TOOL_SCHEMAS_SNAPSHOT = 'tool-schemas.expected.json'

+/** Return the dedicated tool-schema sidecar for one child fixture index. */
+function childToolSchemasSnapshot(index: number): string {
+  return `tool-schemas.${index}.expected.json`
+}
+
 /** The optional full Windows-native stdout transcript. */
 const WINDOWS_STDOUT_SNAPSHOT = 'stdout.expected.windows.jsonl'

@@ -103,6 +108,13 @@ export interface Scenario {
   * declare the same {@link expectedHeaderChanges}; meaningless off a pin.
   */
  toolSchemasSource?: string
+  /**
+   * Child fixture indices whose own schema sequence is pinned separately,
+   * where `1` names `session.1.jsonl` and
+   * `tool-schemas.1.expected.json`. The class pin still owns every other
+   * request-header field.
+   */
+  pinsChildToolSchemas?: readonly number[]
  /**
   * How many changed `request/header` snapshots this PINNING scenario's primary
   * fixture legitimately carries (default 0). Their full prompt text is kept in
@@ -152,25 +164,37 @@ export interface Scenario {
   * test is skipped on Windows; its fixtures stay guarded on every platform.
   */
  posixOnly?: boolean
+  /**
+   * Whether the scenario boots a composition that needs a usable `pwsh`
+   * (the pwsh-tool-turn scenario). The run test is skipped when the suite's
+   * {@link SnapshotSuiteOptions.hasPwsh} probe is false; fixtures stay guarded
+   * on every platform.
+   */
+  pwshOnly?: boolean
 }

 /**
 * Whether a scenario's run test is skipped for this mode and host: record mode
- * skips authored (non-`recorded`) scenarios, and {@link Scenario.posixOnly}
- * scenarios skip on Windows.
+ * skips authored (non-`recorded`) scenarios, {@link Scenario.posixOnly}
+ * scenarios skip on Windows, and {@link Scenario.pwshOnly} scenarios skip
+ * when the caller's `hasPwsh` probe is false.
 *
 * @param scenario The scenario whose run test is being registered.
 * @param recording Whether the suite runs in record mode.
 * @param platform The running Node platform, injectable for unit coverage.
+ * @param hasPwsh The caller's pwsh-availability probe; `pwshOnly` scenarios
+ *   skip unless it is true.
 * @returns True when the scenario's run test must not execute.
 */
 export function scenarioSkipped(
  scenario: Scenario,
  recording: boolean,
  platform: NodeJS.Platform = process.platform,
+  hasPwsh?: boolean,
 ): boolean {
  if (recording && !scenario.recorded) return true
-  return scenario.posixOnly === true && platform === 'win32'
+  if (scenario.posixOnly === true && platform === 'win32') return true
+  return scenario.pwshOnly === true && hasPwsh !== true
 }

 /** One stdout expected output selected for a platform run. */
@@ -211,6 +235,11 @@ export interface SnapshotSuiteOptions {
   * from `$DSH_SNAPSHOT` — env reading stays outside this library.
   */
  mode: 'replay' | 'record' | 'refresh'
+  /**
+   * Whether a real `pwsh` executable is available on this host (the probe the
+   * caller owns; `pwshOnly` scenarios skip when this is not true).
+   */
+  hasPwsh?: boolean
 }

 /** One scenario's generated claim on a shared snapshot file. */
@@ -1052,8 +1081,9 @@ export function defineAcpSnapshotSuite(options: SnapshotSuiteOptions): void {
  scenarioSuite('snapshot scenarios', () => {
    for (const scenario of scenarios) {
      // In RECORD mode, only re-run the `recorded` (live-API) scenarios; the `authored` ones
-      // (sidecar-driven errors/cancel) are never re-recorded. `posixOnly` scenarios skip on Windows.
-      it.skipIf(scenarioSkipped(scenario, RECORDING))(`snapshot: ${scenario.name} matches the expected outputs`, async ({ expect }) => {
+      // (sidecar-driven errors/cancel) are never re-recorded. `posixOnly` scenarios skip on Windows;
+      // `pwshOnly` scenarios skip when the caller's `hasPwsh` probe is false.
+      it.skipIf(scenarioSkipped(scenario, RECORDING, process.platform, options.hasPwsh))(`snapshot: ${scenario.name} matches the expected outputs`, async ({ expect }) => {
        const dir = join(snapshotsDir, scenario.name)
        const input = JSON.parse(await readFile(join(dir, 'input.json'), 'utf8')) as InputScript
        const overrideFile = join(dir, 'replay.override.json')
@@ -1099,6 +1129,8 @@ export function defineAcpSnapshotSuite(options: SnapshotSuiteOptions): void {
          cwdAliases: result.cwdAliases,
        }

+        const childSchemaPins = new Set(scenario.pinsChildToolSchemas ?? [])
+
        // Record writes live model fixtures; keyless refresh writes every comparable replayed
        // fixture. Pinning JSONL keeps prefixes but moves prompts and schemas into sidecars.
        const scrub = scenario.pinsHeader === true
@@ -1177,6 +1209,18 @@ export function defineAcpSnapshotSuite(options: SnapshotSuiteOptions): void {
            claimSharedSnapshot(schemaClaims, schemaPath, scenario.name, toolSchemasSnapshot)
            await writeFile(schemaPath, toolSchemasSnapshot)
          }
+          for (const index of childSchemaPins) {
+            const log = result.sessionLogs[index]
+            expect(log, `${mode}: no child session log at index ${index} to snapshot schemas from`)
+              .toBeDefined()
+            const schemaSets = normalizedToolSchemas((log as HarvestedLog).content, ctx)
+            expect(schemaSets.length, `${mode}: child ${index} produced no tool schemas to snapshot`)
+              .toBeGreaterThan(0)
+            await writeFile(join(dir, childToolSchemasSnapshot(index)), formatToolSchemasSnapshot(
+              schemaSets[0] as unknown[],
+              schemaSets.slice(1),
+            ))
+          }
        }

        for (const expected of stdoutExpectedVariants(scenario)) {
@@ -1230,7 +1274,14 @@ export function defineAcpSnapshotSuite(options: SnapshotSuiteOptions): void {
          header,
          pinnedSchemaSets[index] as unknown[],
        ))
+        const childPinnedSchemas = new Map<number, unknown[][]>()
+        for (const index of childSchemaPins) {
+          const sidecar = await readFile(join(dir, childToolSchemasSnapshot(index)), 'utf8')
+          const parsed = parseToolSchemasSnapshot(sidecar)
+          childPinnedSchemas.set(index, [parsed.initial, ...parsed.changes])
+        }
        for (const [logIndex, log] of result.sessionLogs.entries()) {
+          const childSchemas = childPinnedSchemas.get(logIndex)
          const expectedChanges = scenario.pinsHeader === true && logIndex === 0
            ? scenario.expectedHeaderChanges ?? 0
            : 0
@@ -1243,8 +1294,15 @@ export function defineAcpSnapshotSuite(options: SnapshotSuiteOptions): void {
            .toBe(headers.length)
          expect(schemaSets.length, `session ${log.id}: every request/header must carry an array-valued tools field`)
            .toBe(headers.length)
+          if (childSchemas !== undefined) {
+            expect(childSchemas.length, `session ${log.id}: ${childToolSchemasSnapshot(logIndex)} has an unexpected tool-schema count`)
+              .toBe(schemaSets.length)
+          }
          for (const [k, header] of headers.entries()) {
-            const expected = expectedChanges > 0 ? pinnedHeaders[k] : pinnedHeaders[0]
+            const classPin = expectedChanges > 0 ? pinnedHeaders[k] : pinnedHeaders[0]
+            const expected = childSchemas === undefined
+              ? classPin
+              : { ...classPin as Record<string, unknown>, tools: childSchemas[k] }
            expect(header, `session ${log.id}: request/header #${k + 1} diverged from the pinned (${pinningScenario.name}) header`)
              .toEqual(expected)
            if (expectedChanges === 0) {
@@ -1282,8 +1340,16 @@ export function defineAcpSnapshotSuite(options: SnapshotSuiteOptions): void {

    it('every registered scenario has its required fixture files', async () => {
      // Every scenario needs input, stdout, a primary session fixture, and matching optional sidecars.
-      for (const { name, overridden, pinsNativeWindowsStdout } of scenarios) {
+      for (const { name, overridden, pinsNativeWindowsStdout, pinsChildToolSchemas } of scenarios) {
        const dir = join(snapshotsDir, name)
+        const declaredChildPins = new Set(pinsChildToolSchemas ?? [])
+        const childSidecars = (await readdir(dir, { withFileTypes: true }))
+          .filter(entry => entry.isFile())
+          .map(entry => /^tool-schemas\.([1-9]\d*)\.expected\.json$/.exec(entry.name))
+          .filter((match): match is RegExpExecArray => match !== null)
+          .map(match => Number(match[1]))
+        expect(new Set(childSidecars), `${name}: child tool-schema sidecars must match \`pinsChildToolSchemas\``)
+          .toEqual(declaredChildPins)
        expect(existsSync(join(dir, 'input.json')), `${name}/input.json`).toBe(true)
        expect(existsSync(join(dir, 'stdout.expected.jsonl')), `${name}/stdout.expected.jsonl`).toBe(true)
        expect(
@@ -1366,9 +1432,30 @@ export function defineAcpSnapshotSuite(options: SnapshotSuiteOptions): void {
      assertUniqueSnapshotContents('tool-schema', schemas)
    })

-    it('every committed JSONL has valid tool results and canonical header storage', async () => {
+    it('every declared child tool-schema sidecar is canonical and names a real child', async () => {
+      for (const scenario of scenarios) {
+        const pins = scenario.pinsChildToolSchemas ?? []
+        if (pins.length === 0) continue
+        const dir = join(snapshotsDir, scenario.name)
+        const files = await sessionFixtures(dir)
+        for (const index of pins) {
+          expect(files[index], `${scenario.name}: child schema pin ${index} must name an existing session.<n>.jsonl fixture`)
+            .toBeDefined()
+          const file = childToolSchemasSnapshot(index)
+          const sidecar = await readFile(join(dir, file), 'utf8')
+          const parsed = parseToolSchemasSnapshot(sidecar)
+          expect(sidecar, `${scenario.name}/${file} must use canonical JSON formatting`)
+            .toBe(formatToolSchemasSnapshot(parsed.initial, parsed.changes))
+          expect(parsed.initial.length, `${scenario.name}/${file} must pin at least one schema`)
+            .toBeGreaterThan(0)
+        }
+      }
+    })
+
+    it('every committed JSONL has valid tool results and canonical fixture storage', async () => {
      // Prompts and schemas always leave JSONL. Header pins retain prefixes;
-      // every other fixture tokenizes those too. Fixed-point checks make both
+      // every other fixture tokenizes those too. Portable cwd tokens never
+      // retain a platform realpath prefix. Fixed-point checks make these
      // storage rules fail loud.
      for (const scenario of scenarios) {
        const dir = join(snapshotsDir, scenario.name)
@@ -1377,6 +1464,8 @@ export function defineAcpSnapshotSuite(options: SnapshotSuiteOptions): void {
          const fixture = await readFile(join(dir, file), 'utf8')
          expect(unknownToolCallIds(fixture), `${scenario.name}/${file} contains UNKNOWN_TOOL`)
            .toEqual([])
+          expect(fixture, `${scenario.name}/${file} carries a non-canonical macOS cwd token`)
+            .not.toContain('/private{{cwd}}')
          expect(scrubSystemPrompts(fixture), `${scenario.name}/${file} carries an unscrubbed system prompt`)
            .toEqual(fixture)
          expect(scrubToolSchemas(fixture), `${scenario.name}/${file} carries unscrubbed tool schemas`)
@@ -9,7 +9,7 @@
    ]},
    { "file": "b/child/session.jsonl", "lines": [
      { "type": "session", "id": "eeeeeeee-1111-4222-8333-444444444444", "createdAt": 300, "cwd": "{{CWD}}", "parentSession": "{{SID}}", "delegationDepth": 1 },
-      { "type": "request/header", "seq": 0, "time": 6, "data": { "header": { "config": { "model": "fake" }, "system": "SYS PROMPT", "tools": [{ "name": "t1", "description": "D1", "parameters": { "type": "object" } }] }, "reason": "initial" } }
+      { "type": "request/header", "seq": 0, "time": 6, "data": { "header": { "config": { "model": "fake" }, "system": "SYS PROMPT", "tools": [{ "name": "child-only", "description": "Child D", "parameters": { "type": "object" } }] }, "reason": "initial" } }
    ]}
  ]
 }
@@ -0,0 +1,12 @@
+{
+  "initial": [
+    {
+      "name": "child-only",
+      "description": "Child D",
+      "parameters": {
+        "type": "object"
+      }
+    }
+  ],
+  "changes": []
+}
@@ -584,6 +584,55 @@ describe('runScenario', () => {
    expect(result.sessionLogs[0]?.content).toContain('"type":"turn/end"')
  })

+  it('waitForInboxMessage holds the app through a matching durable insertion', { timeout: 20_000 }, async () => {
+    const { fixtureFile } = await scenario({
+      prompt: 'hang-until-cancel',
+      persistLogsOnCancel: true,
+      logs: [{
+        file: 'project/main/session.jsonl',
+        lines: [
+          { type: 'session', version: 0, id: '{{SID}}', createdAt: 1, delegationDepth: 0 },
+          {
+            type: 'agent/inbox/spliced',
+            seq: 0,
+            time: 2,
+            data: {
+              target: 'next-turn',
+              start: 0,
+              inserted: [{ role: 'user', content: [{ type: 'text', text: 'durable marker' }] }],
+            },
+          },
+        ],
+      }],
+    })
+    const result = await runScenario(
+      { steps: [...boot, { op: 'promptAndCancel', text: 'hang' }, { op: 'waitForInboxMessage', text: 'marker' }] },
+      { agent: AGENT, mode: 'replay', fixtureFile },
+    )
+    expect(result.sessionLogs[0]?.content).toContain('durable marker')
+  })
+
+  it('waitForInboxMessage times out when the session log or matching insertion is absent', { timeout: 20_000 }, async () => {
+    const absent = await scenario({ prompt: 'hang-until-cancel', persistLogsOnCancel: true })
+    await expect(runScenario(
+      { steps: [...boot, { op: 'promptAndCancel', text: 'hang' }, { op: 'waitForInboxMessage', text: 'missing', timeoutMs: 20 }] },
+      { agent: AGENT, mode: 'replay', fixtureFile: absent.fixtureFile },
+    )).rejects.toThrow(/did not persist expected inbox message within 20ms/)
+
+    const unmatched = await scenario({
+      prompt: 'hang-until-cancel',
+      persistLogsOnCancel: true,
+      logs: [{
+        file: 'project/main/session.jsonl',
+        lines: [{ type: 'session', version: 0, id: '{{SID}}', createdAt: 1, delegationDepth: 0 }],
+      }],
+    })
+    await expect(runScenario(
+      { steps: [...boot, { op: 'promptAndCancel', text: 'hang' }, { op: 'waitForInboxMessage', text: 'missing', timeoutMs: 20 }] },
+      { agent: AGENT, mode: 'replay', fixtureFile: unmatched.fixtureFile },
+    )).rejects.toThrow(/did not persist expected inbox message within 20ms/)
+  })
+
  it('waitForTitleAfterTurnEnd holds the app through a standalone durable title', { timeout: 20_000 }, async () => {
    const { fixtureFile } = await scenario({
      prompt: 'hang-until-cancel',
@@ -742,6 +791,125 @@ describe('runScenario', () => {
    )).rejects.toThrow(/did not persist turn\/end within 20ms/)
  })

+  it('waitForGoalPhase requires the requested durable goal phase', { timeout: 20_000 }, async () => {
+    const reached = await scenario({
+      prompt: 'hang-until-cancel',
+      persistLogsOnCancel: true,
+      logs: [{
+        file: 'project/main/session.jsonl',
+        lines: [
+          { type: 'session', version: 0, id: '{{SID}}', createdAt: 1, delegationDepth: 0 },
+          { type: 'goal/change', seq: 1, time: 2, data: {} },
+          { type: 'goal/change', seq: 2, time: 3, data: { goal: { phase: 'active' } } },
+        ],
+      }],
+    })
+    const result = await runScenario(
+      {
+        steps: [
+          ...boot,
+          { op: 'promptAndCancel', text: 'hang' },
+          { op: 'waitForGoalPhase', phase: 'active' },
+        ],
+      },
+      { agent: AGENT, mode: 'replay', fixtureFile: reached.fixtureFile },
+    )
+    expect(result.sessionLogs[0]?.content).toContain('"phase":"active"')
+
+    const missing = await scenario({})
+    await expect(runScenario(
+      { steps: [...boot, { op: 'waitForGoalPhase', phase: 'blocked', timeoutMs: 20 }] },
+      { agent: AGENT, mode: 'replay', fixtureFile: missing.fixtureFile },
+    )).rejects.toThrow(/did not persist goal phase "blocked" within 20ms/)
+  })
+
+  it('waitForSubagentTurnEnd requires a closed child work turn', { timeout: 20_000 }, async () => {
+    const closed = await scenario({
+      prompt: 'hang-until-cancel',
+      persistLogsOnCancel: true,
+      logs: [
+        {
+          file: 'project/main/session.jsonl',
+          lines: [
+            { type: 'session', version: 0, id: '{{SID}}', createdAt: 1, delegationDepth: 0 },
+            { type: 'turn/end', seq: 1, time: 2, data: { turn: 1, reason: { kind: 'aborted' } } },
+          ],
+        },
+        {
+          file: 'project/child/session.jsonl',
+          lines: [
+            { type: 'session', version: 0, id: 'child-1', createdAt: 2, parentSession: '{{SID}}', delegationDepth: 1 },
+            { type: 'subagent/descriptor', seq: 0, time: 1, data: {} },
+            { type: 'turn/start', seq: 1, time: 2, data: { turn: 1 } },
+            { type: 'request/header', seq: 2, time: 3, data: { header: {}, reason: 'initial' } },
+            { type: 'turn/end', seq: 3, time: 4, data: { turn: 1, reason: { kind: 'completed' } } },
+          ],
+        },
+      ],
+    })
+    const result = await runScenario(
+      {
+        steps: [
+          ...boot,
+          { op: 'promptAndCancel', text: 'hang' },
+          { op: 'waitForSubagentTurnEnd' },
+        ],
+      },
+      { agent: AGENT, mode: 'replay', fixtureFile: closed.fixtureFile },
+    )
+    expect(result.sessionLogs[1]?.parentSession).toBe(result.sessionId)
+    await expect(runScenario(
+      {
+        steps: [
+          ...boot,
+          { op: 'promptAndCancel', text: 'hang' },
+          { op: 'waitForSubagentTurnEnd', minimumTurn: 2, timeoutMs: 20 },
+        ],
+      },
+      { agent: AGENT, mode: 'replay', fixtureFile: closed.fixtureFile },
+    )).rejects.toThrow(/subagent child #1 did not persist closed turn 2 within 20ms/)
+
+    const seedOnly = await scenario({
+      prompt: 'hang-until-cancel',
+      persistLogsOnCancel: true,
+      logs: [
+        {
+          file: 'project/main/session.jsonl',
+          lines: [
+            { type: 'session', version: 0, id: '{{SID}}', createdAt: 1, delegationDepth: 0 },
+            { type: 'turn/end', seq: 1, time: 2, data: { turn: 1, reason: { kind: 'aborted' } } },
+          ],
+        },
+        {
+          file: 'project/child/session.jsonl',
+          lines: [
+            { type: 'session', version: 0, id: 'child-1', createdAt: 2, parentSession: '{{SID}}', delegationDepth: 1 },
+            { type: 'turn/start', seq: 0, time: 1, data: { turn: 1, trigger: { kind: 'message' } } },
+            { type: 'request/header', seq: 1, time: 2, data: { header: {}, reason: 'initial' } },
+            { type: 'turn/end', seq: 2, time: 3, data: { turn: 1, reason: { kind: 'completed' } } },
+            { type: 'subagent/descriptor', seq: 3, time: 4, data: {} },
+          ],
+        },
+      ],
+    })
+    await expect(runScenario(
+      {
+        steps: [
+          ...boot,
+          { op: 'promptAndCancel', text: 'hang' },
+          { op: 'waitForSubagentTurnEnd', timeoutMs: 20 },
+        ],
+      },
+      { agent: AGENT, mode: 'replay', fixtureFile: seedOnly.fixtureFile },
+    )).rejects.toThrow(/subagent child #1 did not persist closed turn 1 within 20ms/)
+
+    const missing = await scenario({})
+    await expect(runScenario(
+      { steps: [...boot, { op: 'waitForSubagentTurnEnd', child: 2, timeoutMs: 20 }] },
+      { agent: AGENT, mode: 'replay', fixtureFile: missing.fixtureFile },
+    )).rejects.toThrow(/subagent child #2 did not persist closed turn 1 within 20ms/)
+  })
+
  it('waitForTitleAfterTurnEnd times out when the title precedes the boundary', { timeout: 20_000 }, async () => {
    const { fixtureFile } = await scenario({
      prompt: 'hang-until-cancel',
@@ -767,6 +935,55 @@ describe('runScenario', () => {
    )).rejects.toThrow(/did not persist session\/title after turn\/end within 20ms/)
  })

+  it('waitForEventAfterTurnEnd holds the app for a typed post-boundary record and times out otherwise', { timeout: 20_000 }, async () => {
+    const late = await scenario({
+      prompt: 'hang-until-cancel',
+      persistLogsOnCancel: true,
+      logs: [{
+        file: 'project/main/session.jsonl',
+        lines: [
+          { type: 'session', version: 0, id: '{{SID}}', createdAt: 1, delegationDepth: 0 },
+          { type: 'turn/end', seq: 1, time: 2, data: { turn: 1, reason: { kind: 'aborted' } } },
+          { type: 'user/message', seq: 2, time: 3, data: { content: [{ type: 'text', text: 'late goal state' }], source: { kind: 'user' } } },
+        ],
+      }],
+    })
+    const result = await runScenario(
+      {
+        steps: [
+          ...boot,
+          { op: 'promptAndCancel', text: 'hang' },
+          { op: 'waitForEventAfterTurnEnd', type: 'user/message' },
+        ],
+      },
+      { agent: AGENT, mode: 'replay', fixtureFile: late.fixtureFile },
+    )
+    expect(result.sessionLogs[0]?.content).toMatch(/"turn\/end"[\s\S]*"user\/message"/)
+
+    const early = await scenario({
+      prompt: 'hang-until-cancel',
+      persistLogsOnCancel: true,
+      logs: [{
+        file: 'project/main/session.jsonl',
+        lines: [
+          { type: 'session', version: 0, id: '{{SID}}', createdAt: 1, delegationDepth: 0 },
+          { type: 'user/message', seq: 1, time: 1, data: { content: [{ type: 'text', text: 'early' }], source: { kind: 'user' } } },
+          { type: 'turn/end', seq: 2, time: 2, data: { turn: 1, reason: { kind: 'aborted' } } },
+        ],
+      }],
+    })
+    await expect(runScenario(
+      {
+        steps: [
+          ...boot,
+          { op: 'promptAndCancel', text: 'hang' },
+          { op: 'waitForEventAfterTurnEnd', type: 'user/message', timeoutMs: 20 },
+        ],
+      },
+      { agent: AGENT, mode: 'replay', fixtureFile: early.fixtureFile },
+    )).rejects.toThrow(/did not persist user\/message after turn\/end within 20ms/)
+  })
+
  it('promptExpectError swallows a model-error response as the expected outcome', { timeout: 20_000 }, async () => {
    const { fixtureFile } = await scenario({ prompt: 'error' })
    const result = await runScenario(
@@ -865,6 +1082,19 @@ describe('runScenario', () => {
    expect(result.sessionId).toBeDefined()
  })

+  it('waitForFile holds the next input step behind cwd-relative readiness', { timeout: 20_000 }, async () => {
+    const { dir, fixtureFile } = await scenario({})
+    const workspaceDir = join(dir, 'workspace')
+    const { mkdir } = await import('node:fs/promises')
+    await mkdir(workspaceDir, { recursive: true })
+    await writeFile(join(workspaceDir, 'ready'), '')
+    const result = await runScenario(
+      { steps: [...boot, { op: 'waitForFile', path: 'ready' }, { op: 'cancel' }] },
+      { agent: AGENT, mode: 'replay', fixtureFile, workspaceDir },
+    )
+    expect(result.sessionId).toBeDefined()
+  })
+
  it.each([
    [{ op: 'prompt', text: 'x' }, /prompt before newSession/],
    [{ op: 'promptAndWaitForAgentMessage', text: 'x', waitForText: 'later' }, /promptAndWaitForAgentMessage before newSession/],
@@ -872,7 +1102,10 @@ describe('runScenario', () => {
    [{ op: 'promptAndCancel', text: 'x' }, /promptAndCancel before newSession/],
    [{ op: 'waitForTurnStart' }, /waitForTurnStart before newSession/],
    [{ op: 'waitForTurnEnd' }, /waitForTurnEnd before newSession/],
+    [{ op: 'waitForGoalPhase', phase: 'active' }, /waitForGoalPhase before newSession/],
+    [{ op: 'waitForInboxMessage', text: 'marker' }, /waitForInboxMessage before newSession/],
    [{ op: 'waitForTitleAfterTurnEnd' }, /waitForTitleAfterTurnEnd before newSession/],
+    [{ op: 'waitForEventAfterTurnEnd', type: 'user/message' }, /waitForEventAfterTurnEnd before newSession/],
    [{ op: 'cancel' }, /cancel before newSession/],
  ] as [InputStep, RegExp][])('rejects %j before newSession', { timeout: 20_000 }, async (step, message) => {
    const { fixtureFile } = await scenario({})
@@ -470,6 +470,24 @@ describe('tokenizeSessionFixtureCwd', () => {
    expect(tokenizeSessionFixtureCwd(out)).toBe(out)
  })

+  it('collapses a residual macOS realpath prefix around an existing cwd token', () => {
+    const raw = [
+      JSON.stringify({ type: 'session', id: 's', createdAt: 1, cwd: '{{cwd}}' }),
+      JSON.stringify({
+        type: 'tool/result',
+        seq: 1,
+        time: 2,
+        data: { content: [{ type: 'text', text: 'wrote /private{{cwd}}/proof.txt' }] },
+      }),
+      '',
+    ].join('\n')
+
+    const out = tokenizeSessionFixtureCwd(raw)
+    expect(out).toContain('wrote {{cwd}}/proof.txt')
+    expect(out).not.toContain('/private{{cwd}}')
+    expect(tokenizeSessionFixtureCwd(out)).toBe(out)
+  })
+
  it('rejects a log without a session cwd', () => {
    expect(() => tokenizeSessionFixtureCwd('')).toThrow(
      'acp-snapshot: cannot tokenize a cwd without a basename',
@@ -83,6 +83,7 @@ const REPLAY_SCENARIOS: Scenario[] = [
    env: { DSH_PERMISSION_MODE: 'never' },
    configPath: AGENT.configPath,
    workspaceParent: tmpdir(),
+    pinsChildToolSchemas: [1],
    prepareWorkspace: (cwd) => {
      writeFileSync(join(cwd, 'seed.txt'), 'prepared at runtime')
    },
@@ -94,7 +95,7 @@ const REPLAY_SCENARIOS: Scenario[] = [

 const RECORD_SCENARIOS: Scenario[] = [
  { name: 'rec-pin', hasModelTurn: true, recorded: true, pinsHeader: true },
-  { name: 'rec-child', hasModelTurn: true, recorded: true },
+  { name: 'rec-child', hasModelTurn: true, recorded: true, pinsChildToolSchemas: [1] },
  // recorded:false in record mode → registered but skipped (never re-recorded).
  { name: 'rec-skip', hasModelTurn: true, recorded: false, overridden: true },
 ]
@@ -123,6 +124,7 @@ function staleRefreshFixtures(dir: string): void {
  writeFileSync(join(dir, 'plain-turn', 'stdout.expected.jsonl'), 'stale stdout\n')
  writeFileSync(join(dir, 'pin-turn', 'system-prompt.expected.md'), 'STALE PROMPT\n')
  writeFileSync(join(dir, 'pin-turn', 'tool-schemas.expected.json'), '{"initial":[{"name":"stale"}],"changes":[]}\n')
+  writeFileSync(join(dir, 'plain-turn', 'tool-schemas.1.expected.json'), '{"initial":[{"name":"stale-child"}],"changes":[]}\n')

  const plainBehaviorFile = join(dir, 'plain-turn', 'behavior.json')
  const plainBehavior = JSON.parse(readFileSync(plainBehaviorFile, 'utf8')) as Record<string, unknown>
@@ -185,6 +187,9 @@ describe('defineAcpSnapshotSuite: refresh write-back', () => {
    const schemas = readFileSync(join(refreshDir, 'pin-turn', 'tool-schemas.expected.json'), 'utf8')
    expect(schemas).toContain('"description": "D1"')
    expect(schemas).not.toContain('"name":"stale"')
+    const childSchemas = readFileSync(join(refreshDir, 'plain-turn', 'tool-schemas.1.expected.json'), 'utf8')
+    expect(childSchemas).toContain('"name": "child-only"')
+    expect(childSchemas).not.toContain('stale-child')

    const pinSession = readFileSync(join(refreshDir, 'pin-turn', 'session.jsonl'), 'utf8')
    expect(pinSession).toContain('"cwd":"{{cwd}}"')
@@ -197,6 +202,8 @@ describe('defineAcpSnapshotSuite: record inventory write-back', () => {
    expect(fixture).toContain('"type":"session"')
    expect(fixture).toContain('"cwd":"{{cwd}}"')
    expect(() => readFileSync(join(recordDir, 'rec-child', 'session.2.jsonl'), 'utf8')).toThrow()
+    expect(readFileSync(join(recordDir, 'rec-child', 'tool-schemas.1.expected.json'), 'utf8'))
+      .toContain('"name": "t1"')
  })

  it('retains an unchanged message id across the recorded parent and child fixtures', () => {
@@ -469,6 +476,7 @@ describe('stdoutExpectedVariants', () => {
 describe('scenarioSkipped', () => {
  const authored: Scenario = { name: 'authored', hasModelTurn: true, recorded: false }
  const posix: Scenario = { name: 'posix-cancel', hasModelTurn: true, recorded: false, posixOnly: true }
+  const pwsh: Scenario = { name: 'pwsh-tool', hasModelTurn: true, recorded: false, pwshOnly: true }

  it('skips authored scenarios only while recording', () => {
    expect(scenarioSkipped(authored, true, 'linux')).toBe(true)
@@ -481,6 +489,13 @@ describe('scenarioSkipped', () => {
    expect(scenarioSkipped(posix, false, 'darwin')).toBe(false)
    expect(scenarioSkipped(authored, false, 'win32')).toBe(false)
  })
+
+  it('skips pwshOnly scenarios when the host lacks pwsh, and runs them otherwise', () => {
+    expect(scenarioSkipped(pwsh, false, 'linux', false)).toBe(true)
+    expect(scenarioSkipped(pwsh, false, 'win32', true)).toBe(false)
+    expect(scenarioSkipped(pwsh, false, 'linux', true)).toBe(false)
+    expect(scenarioSkipped(authored, false, 'linux', false)).toBe(false)
+  })
 })

 describe('fixtureContext', () => {