$ loading_
在每次 PR 变更时评估文档影响并生成可执行的更新建议
复制安装指令，让 AI 自动完成配置 · 推荐新手
请帮我安装 askskill 上的 "docs-sync" 技能： 1. 下载 https://raw.githubusercontent.com/microsoft/apm/main/.agents/skills/docs-sync/SKILL.md 2. 保存为 ~/.claude/skills/docs-sync/SKILL.md 3. 装好后重载技能，告诉我可以用了
请检查这个 PR 的代码变更，判断是否会导致现有文档失真，并将影响分类为无需修改、局部改写或结构性调整；给出需要更新的页面、原因和补丁级建议。
一份文档影响评估，包含影响等级、受影响文档位置、建议修改内容和后续协作建议。
即使 PR 标题和描述没有提到文档，也请基于代码差异检查是否存在文档漂移，并指出哪些说明、示例或导航结构会因此过时。
指出代码与文档不一致的具体位置，并说明是文字更新还是需要新增页面或调整目录。
根据这个 PR 输出一份可直接交给文档团队的建议单，包含变更摘要、受影响读者、推荐修改步骤、页面优先级，以及不在本次范围内的事项。
一份结构清晰的文档更新建议单，便于作者和编辑快速据此落地修改。
The docs corpus drifts silently and constantly. This skill catches drift at PR-open time, classifies its impact, and orchestrates a persona panel to produce a patch-ready advisory comment.
The pattern is A1 PANEL + B1 FAN-OUT/SYNTHESIZER + A8 ALIGNMENT LOOP. The classifier is the cost gate (~70% of PRs short-circuit to no-change with ~1 LLM call). When the panel does fan out, every agent reads a bounded context (~10 KB) -- never the full corpus.
This skill is ADVISORY. It does not gate merge, apply verdict labels, or push to the contributor's fork. The orchestrator is the sole writer to the PR: exactly one comment per run (idempotent edit-in-place), plus optional label sweeps.
N/15 for observability.gh write command, post comments, or touch PR state.## Docs sync advisory. Re-runs edit-in-place using gh pr comment --edit-last.docs-sync-confirm label) open from a bot branch in the BASE repo; never pushed to the contributor's fork..apm/docs-index.ymlapm --help, grep, and python -c commands to verify doc claims, never assert from prose.| Role | Agent | Always active? |
|---|---|---|
| Classifier | doc-analyser inside docs-impact-classifier | Yes (every run) |
| Localizer | docs-impact-localizer | Only on in_place verdict |
| Architect | docs-impact-architect | Only on structural verdict |
| Writer | doc-writer | Per candidate page (fan-out) |
| Verifier | python-architect | Per candidate page (fan-out, S7) |
| Editorial | editorial-owner | Once across all redrafts |
| Growth | oss-growth-hacker | Once across all redrafts |
| Synthesizer | cdo | Once, with ALIGNMENT LOOP up to 3 redrafts |
docs-sync SKILL (orchestrator thread)
|
Step 1: classify (1 LLM call, may exit here)
|
v
verdict?
/ | \
no-change in-place structural
| | |
EXIT | architect (TOC delta)
| |
+----<-----+
|
Step 2: localize (1 LLM call) -- per-page task brief
|
Step 3: FAN-OUT panel via task tool
|
+----+----+----+----+
v v v v v
writer verify edit growth
x N x N once once
(parallel; each <=10 KB context)
|
Step 4: schema-validate returns
|
Step 5: CDO synthesize (1 LLM call)
|
agree?
/ | \
revise (N<=3 redrafts) | agree
|
Step 6: emit ONE comment via safe-outputs.add-comment
Step 7: OPTIONAL companion docs PR (only if structural AND
`docs-sync-confirm` label present)
Spawn ONE task: load the docs-impact-classifier skill, pass it the
…
优化 APM 代码库中的 CLI 输出、日志提示与诊断信息体验
在文档结构需随 PR 调整时，规划目录变更并生成新页面大纲草案。