skills/sync-release-docs/SKILL.md
Sync project documentation after release. Reads all docs, cross-references the diff, updates README/ARCHITECTURE/CONTRIBUTING/CLAUDE.md to match what shipped, polishes CHANGELOG voice, cleans up TODOS. Use when asked to "update the docs", "sync documentation", or "post-ship docs".
npx skillsauth add nesnilnehc/ai-cortex sync-release-docsInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
在代码发版或 PR 合并后,确保项目文档(README、ARCHITECTURE、CONTRIBUTING、CLAUDE.md、CHANGELOG、TODOS)与已交付变更一致,保持可发现性与跨文档一致性。
首要目标:产出与发版变更一致的文档更新;事实性修正自动完成,主观或风险性变更经用户确认。
成功标准(必须满足所有要求):
.ai-cortex/config.yaml 或 gh 推断 base branch;参见 docs/guides/project-config.mdgit diff、git log 已执行;README、ARCHITECTURE、CONTRIBUTING、CLAUDE.md 已对照 diff 核对;分类为 Auto-update 或 Ask user验收测试:新贡献者能否仅凭 README/CONTRIBUTING 正确完成首次贡献流程?
本技能负责:
本技能不负责:
discover-docs-norms)assess-docs)bootstrap-docs)转交点:文档同步完成后,移交给后续发版流程或下一个任务。若发现规范违规,建议运行 assess-docs。
CLAUDE.md 或 .ai-cortex/config.yaml,优先读取 base_branchgh pr view --json baseRefName -q .baseRefName(若有 PR);失败则 gh repo view --json defaultBranchRef -q .defaultBranchRef.name;均失败则 maingit diff <base>...HEAD --stat、git log <base>..HEAD --oneline、git diff <base>...HEAD --name-onlyfind . -maxdepth 2 -name "*.md" -not -path "./.git/*" -not -path "./node_modules/*" | sort对 README、ARCHITECTURE、CONTRIBUTING、CLAUDE.md 及项目内其他 .md:
每项更新分类为 Auto-update(事实性、diff 明确)或 Ask user(叙述性、大面积重写、安全模型、歧义)。
对明确的事实性更新直接使用 Edit 工具修改。每处修改输出一行摘要(具体改了什么)。
禁止自动更新:README 定位、ARCHITECTURE 设计理由、安全模型、删除整节。
对 Step 2 标记为 Ask user 的项,使用 AskUserQuestion,附 RECOMMENDATION 与选项(含 Skip)。用户确认后立即应用。
禁止重写或替换 CHANGELOG 条目。
若存在 TODOS.md:将 diff 明确完成的项移至 Completed;保守标记。可选:检查 diff 中的 TODO/FIXME 是否应进入 TODOS。
禁止擅自 bump VERSION。
git add <files>、git commit -m "docs: sync documentation for release"、输出 doc health summaryDoc health summary 格式:
Documentation health:
README.md [Updated|Current] (description)
ARCHITECTURE.md [Updated|Current|—] (description)
CONTRIBUTING.md [Updated|Current|—] (description)
CHANGELOG.md [Voice polished|Current|—] (description)
TODOS.md [Updated|Current|—] (description)
VERSION [Not bumped|Already bumped|Skipped] (description)
gh 推断不要做这些(其他技能可以处理它们):
discover-docs-normsassess-docsbootstrap-docs新贡献者能否仅凭 README/CONTRIBUTING 完成首次贡献?
场景:新技能 sync-release-docs 已合并,需更新 README 能力表与 CHANGELOG。
步骤:检测 base branch → 分析 diff → 发现 README 缺新技能、CHANGELOG 已有条目 → Auto-update README 表 → 润色 CHANGELOG 措辞 → 确认不 bump VERSION → commit。
场景:文档已与 diff 一致。
输出:All documentation is up to date.
本技能产出更新后的文档与单次 commit:
| 元素 | 格式 | 必填字段 | 路径模式 | | :--- | :--- | :--- | :--- | | 文档变更 | 文件 diff | 涉及 README / ARCHITECTURE / CONTRIBUTING / CLAUDE.md / CHANGELOG / TODOs;每项含 reason | 项目根及 docs/ 下相应文件 | | 提交 | git commit | 单次 commit;message 遵循 Conventional Commits;body 列出 synced 文件清单 | git history | | 健康摘要 | 结构化聊天输出 | 字段:files_updated[]、files_skipped[](含理由)、todos_resolved[]、commit_hash | 标准输出 |
development
Generate an LLM agent test suite (golden cases, mock-LLM unit tests, evaluator harness) from an agent implementation and its agent-test contract. Use when an agent has no tests, or a contract exists but the test code is missing.
development
After code changes, auto-detect the project's build system and local deployment method for a given directory, then build the project and restart its locally-deployed environment (Docker Compose / systemd / process manager). Never assumes — asks only when detection is ambiguous. Caches detected commands per project in .cortex/redeploy-local.yaml; re-invocations on the same project skip re-scanning until signal files change, the cache expires (30 days), or the skill version bumps.
tools
Publish a NATS message conforming to a cross-team contract, using NATS MCP tools. Authors the contract on first use if missing. Reads project-level cache (.cortex/nats.yaml) to avoid re-prompting basics across sessions.
tools
Drain pending NATS messages from a producer contract via NATS MCP tools (default batch / drain-style). Applies Tolerant Reader semantics and per-message ack/nak/term, returning aggregated stats. Reads project-level cache (.cortex/nats.yaml) to avoid re-prompting.