docs: purge chain-of-thought leakage from prose
Delete design-session citations (decision/audit/plan ordinals, stack positions), change narration, review choreography, and reviewer-addressed justification from comments, JSDoc, docs, READMEs, Agent Notes, tests, and generator templates; restate every affected fact as current-state contract prose. Fix generated docs at their sources and regenerate the catalogs and cordis-surface regions; re-paste type-equiv blocks; update every bilingual counterpart and re-record the pairs. Record the citation rule in the committed-artifact-citations Agent Note.
This commit is contained in:
Tianyi Cui
@@ -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: 0b935ef60c33fd24660d8ecf2497f5506157c724
|
||||
README.zh.md: 7c1c3328ab8490c7eebd4b2492bc4ae6f4fcd5a5
|
||||
README.md: 8a9669b1f555fceb7af6d017d273b62d7dcad1e1
|
||||
README.zh.md: efa31df86d3b494ab14fc3742fc84adce570a593
|
||||
|
||||
@@ -11,7 +11,7 @@ Four layers, importable separately:
|
||||
- **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, mutually unique messages by structurally rewriting only complete surface and durable-inbox message ID fields 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 only when both its ID and identity-free fingerprint are unique across the scenario's fixture-ready parent/child logs; the session package's authoritative surface-type predicate selects surface carriers, correlated `agent/inbox/spliced` copies join the same mapping, and only validated `id` fields in those carriers are rewritten. New, changed, malformed, and graph-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; complete message IDs in surface or inbox carriers are excluded because the later structural pass owns them, 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.
|
||||
Committed session fixtures use canonical packed rows; the [temporary repository migrator](../../../scripts/migrate-packed-session-fixtures.ts) (`pnpm run migrate:packed-session-fixtures`) rewrites older fixture layouts, and its [removal proposal](../../../.agents/notes/proposed/process/2026-07-26-remove-packed-session-fixture-migrator.md) owns its deletion.
|
||||
|
||||
A consuming `*.snapshot.ts` is the scenario table plus one factory call:
|
||||
|
||||
|
||||
@@ -11,7 +11,7 @@ ACP 快照套件工具包：无密钥快照层（`pnpm run test:snapshot`，见[
|
||||
- **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 的每种原生／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，包括已进行 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，通过结构化方式仅改写 surface 和持久 inbox 中完整消息的 ID 字段，将已提交 UUID 带入未变化且双向唯一匹配的消息）。
|
||||
- **`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。在录制或刷新模式写入 fixture 前，仅当一条未变化完整消息的 ID 及其去除身份后的指纹在场景可写入 fixture 的父级/子级日志中均唯一时，该消息才会保留已提交的 UUID；会话包的权威 surface 类型谓词负责选择 surface 载体，与其关联的 `agent/inbox/spliced` 副本也纳入同一映射，且仅改写这些载体中通过验证的 `id` 字段。新增、发生变化、格式错误以及图关系存在歧义的消息保留本次生成的 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，才会复用归一化后等价的叶值；surface 或 inbox 载体中的完整消息 ID 不参与此路径，因为后续结构化处理负责这些 ID；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` 同级文件构成有序的主会话／子会话清单；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。合并此约定的在途分支通过 `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。
|
||||
签入仓库的会话 fixture 使用规范打包行；[临时仓库迁移器](../../../scripts/migrate-packed-session-fixtures.ts)（`pnpm run migrate:packed-session-fixtures`）会改写较旧的 fixture 布局，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。
|
||||
|
||||
消费方 `*.snapshot.ts` 就是场景表加一次工厂调用：
|
||||
|
||||
|
||||
@@ -760,8 +760,8 @@ async function harvestSessionLogs(root: string): Promise<HarvestedLog[]> {
|
||||
})
|
||||
}
|
||||
// Primary (no parentSession) first, then children by ascending createdAt. A
|
||||
// scenario has exactly one top-level session. In the synchronous cut sibling
|
||||
// children are created strictly sequentially, so their createdAt values are
|
||||
// scenario has exactly one top-level session. Subagent children are created
|
||||
// synchronously and strictly sequentially, so their createdAt values are
|
||||
// strictly ordered; the recordedId tiebreak only keeps a degenerate
|
||||
// same-millisecond collision (unreachable here) deterministic. This harvest
|
||||
// order must match the replay load order in dsh-llm-replay's loadSessionScripts
|
||||
|
||||
@@ -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/llm-replay/README.md
|
||||
README.md: a6c087778b8e64124590fb6be7a652b58e1b6343
|
||||
README.zh.md: 05dcf994962ba37b58cfb18a68f0d3819d34db0a
|
||||
README.md: ae7cb5414a28bcf33f9878f6f02220861df9a0d5
|
||||
README.zh.md: 2acc0ff8e8c011126452ba3458aaeb0800d9d8f3
|
||||
|
||||
@@ -20,7 +20,7 @@ A scripted string may embed `{{fromRequest:<regex>}}` to fill a value no static
|
||||
|
||||
A scenario where a parent agent delegates to in-process subagents records more than one log: the parent (`session.jsonl`) plus one per child (`session.1.jsonl`, …). Each agent runs as its own `Session` on the same context, so replay must serve each one its own script.
|
||||
|
||||
Replay keys every call by its calling session id (`GenerateOptions.sessionId`, stamped by the agent loop). Live session ids are freshly random each run and never equal the recorded ones, so a live session binds to a recorded script by **first-call order**: scripts are ordered by header `createdAt` (parent first — it streams before it can delegate), and the first live session to make any call claims the first script, the next new session the next, and so on. Each session then advances its own cursor. A call with no `sessionId` is one anonymous session bound to the primary script, so single-session scenarios behave exactly as before. More distinct live sessions than recorded scripts fails loud.
|
||||
Replay keys every call by its calling session id (`GenerateOptions.sessionId`, stamped by the agent loop). Live session ids are freshly random each run and never equal the recorded ones, so a live session binds to a recorded script by **first-call order**: scripts are ordered by header `createdAt` (parent first — it streams before it can delegate), and the first live session to make any call claims the first script, the next new session the next, and so on. Each session then advances its own cursor. A call with no `sessionId` is one anonymous session bound to the primary script. More distinct live sessions than recorded scripts fails loud.
|
||||
|
||||
## Config
|
||||
|
||||
|
||||
@@ -20,7 +20,7 @@ fixture 就是持久化的会话日志（`<scenario>/session.jsonl`）。其 `as
|
||||
|
||||
父 agent 委托给进程内 subagent 的场景会记录多个日志：父会话使用 `session.jsonl`，每个子会话各使用一个日志（`session.1.jsonl` 等）。每个 agent 都在同一上下文中作为独立的 `Session` 运行，因此回放必须为每个 agent 提供各自的脚本。
|
||||
|
||||
回放根据发起调用的会话 id 为每次调用建立键（`GenerateOptions.sessionId` 由 agent loop 写入）。实时会话 id 每次运行时都会重新随机生成，绝不会等于记录中的 id，因此实时会话按**首次调用顺序**绑定到已记录脚本：脚本按 header 中的 `createdAt` 排序（父会话在前，因为它必须先开始流式输出才能委托）；第一个发起调用的实时会话取得第一个脚本，下一个新会话取得下一个脚本，以此类推。此后每个会话分别推进自己的游标。没有 `sessionId` 的调用视为一个绑定主脚本的匿名会话，因此单会话场景的行为与以前完全相同。不同实时会话的数量超过已记录脚本数时会明确报错。
|
||||
回放根据发起调用的会话 id 为每次调用建立键（`GenerateOptions.sessionId` 由 agent loop 写入）。实时会话 id 每次运行时都会重新随机生成，绝不会等于记录中的 id，因此实时会话按**首次调用顺序**绑定到已记录脚本：脚本按 header 中的 `createdAt` 排序（父会话在前，因为它必须先开始流式输出才能委托）；第一个发起调用的实时会话取得第一个脚本，下一个新会话取得下一个脚本，以此类推。此后每个会话分别推进自己的游标。没有 `sessionId` 的调用视为一个绑定主脚本的匿名会话。不同实时会话的数量超过已记录脚本数时会明确报错。
|
||||
|
||||
## 配置
|
||||
|
||||
|
||||
Reference in New Issue
Block a user