跳到正文
cccskills
免费在 GitHub 公开

tdmcp-docs-roadmap-update

Updates tdmcp docs and roadmap for newly shipped or newly merged capabilities. Use when asked to update docs, update the roadmap, reconcile release docs, document new features, sync docs with CHANGELOG/package release, refresh cookbook examples after a feature wave, re-run/continue/fix a docs-roadmap pass, or check whether docs/ROADMAP.md is current. Runs a sub-agent workflow: release audit -> roadmap/docs edit + cookbook sync -> QA.

中文简介正在准备中。当前显示原文说明。

查看安装方法

包含的文件(1)

  • SKILL.md4.4 KB

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

tdmcp-docs-roadmap-update

Updates the repo's documentation surfaces after feature waves or releases: docs/ROADMAP.md, high-level docs named by the audit, EN/PT prompt cookbooks, and generated reference docs.

Execution mode: sub-agents. This environment does not use TeamCreate; spawn the role agents defined in .Codex/agents/ and coordinate them from the main session.

Phase 0: Context Check

Before spawning agents:

  1. Inspect git status --short.
  2. Check whether _workspace/docs_release_audit.md, _workspace/roadmap_docs_update.md, _workspace/cookbook_sync_update.md, or _workspace/docs_roadmap_qa.md already exist.
  3. Decide execution mode:
    • no prior workspace files -> full run;
    • prior files + user says update/re-run/continue -> refresh only affected phases;
    • prior files + new release/tag/feature input -> refresh audit first, then re-run writers.

Preserve unrelated local changes. Do not reset or revert user edits.

Phase 1: Release And Feature Audit

Spawn docs-release-auditor:

Read .Codex/agents/docs-release-auditor.md and execute it.
Audit current local package, CHANGELOG, ROADMAP, README, cookbook docs,
CLI/resource surfaces, generated docs, npm package version, GitHub releases, and
tags. Write _workspace/docs_release_audit.md.

The audit is mandatory whenever release wording, tool counts, or "latest" features are involved.

Phase 2: Docs Writing

After _workspace/docs_release_audit.md exists, spawn these agents in parallel when their write sets are both needed:

Roadmap/docs editor

Read .Codex/agents/roadmap-docs-editor.md and execute it.
Use _workspace/docs_release_audit.md. Update docs/ROADMAP.md and any high-level
docs named by the audit. Do not edit generated docs or cookbook files. Write
_workspace/roadmap_docs_update.md.

Cookbook sync

Read .Codex/agents/docs-cookbook-sync.md and execute it.
Use _workspace/docs_release_audit.md and _workspace/roadmap_docs_update.md if it
exists. Update docs/guide/prompt-cookbook.md and docs/pt/guide/prompt-cookbook.md
for docs-worthy new features, keeping locales in parity. Write
_workspace/cookbook_sync_update.md.

If only roadmap docs or only cookbook docs are affected, run only the relevant writer.

Phase 3: Generated Docs And QA

Run the docs generation/build gate from the main session:

npm run docs:build
npm run validate:recipes
git diff --check

Then spawn docs-roadmap-qa:

Read .Codex/agents/docs-roadmap-qa.md and execute it.
Use the release audit and writer summaries. Verify release claims, EN/PT parity,
generated docs behavior, VitePress media embeds, and command outcomes. Write
_workspace/docs_roadmap_qa.md.

If QA finds narrow issues, fix them in the main session or re-run only the responsible writer.

Data Flow

  • Auditor writes _workspace/docs_release_audit.md.
  • Roadmap editor writes _workspace/roadmap_docs_update.md.
  • Cookbook sync writes _workspace/cookbook_sync_update.md.
  • QA writes _workspace/docs_roadmap_qa.md.
  • Final docs edits remain in normal repo files.

Quality Rules

  • Verify live release state before claiming a public version is current.
  • Keep public release, local HEAD, and planned backlog distinct.
  • Never hand-edit generated docs/reference/tools.md.
  • Keep EN/PT cookbook examples in parity.
  • Use withBase('/examples/...') for VitePress video embeds.
  • Treat missing media assets as description-only or explicit follow-up, not as silent broken links.

Test Scenarios

Normal flow: a new release lands. Audit confirms package/npm/GitHub agree, roadmap is updated with shipped and remaining work, cookbook gets matching EN/PT entries, npm run docs:build, npm run validate:recipes, and git diff --check pass.

Error flow: npm or GitHub release checks fail. Auditor marks release state unverified; writers avoid definitive public-release claims; QA records residual risk and asks for a retry or human release confirmation.

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

同一仓库的技能

Use when manage multiple local CLI agents via tmux sessions (start/stop/monitor/assign) with cron-friendly scheduling.

中文简介正在准备中。当前显示原文说明。

hybridlabor-api/aos62026年10月16日 更新

Use when a hybrid memory system that provides persistent, searchable knowledge management for AI agents (Architecture, Patterns, Decisions).

中文简介正在准备中。当前显示原文说明。

hybridlabor-api/aos62026年10月16日 更新

Reference for how BDB structures autonomous software engineering work — the seven-node dispatcher graph (Architect, TechLead, UI/UX, Engineering, Media/EventTech, Reviewer, Shipping) that /startcycle-graph actually runs. Use when you need the high-level lifecycle framing without inventing your own process.

中文简介正在准备中。当前显示原文说明。

hybridlabor-api/aos62026年10月16日 更新

Tools are how AI agents interact with the world. A well-designed tool is the difference between an agent that works and one that hallucinates, fails silently, or costs 10x more tokens than necessary. This skill covers tool design from schema to error handling.

中文简介正在准备中。当前显示原文说明。

hybridlabor-api/aos62026年10月16日 更新

Harness patterns for coding agents — memory, permissions, context engineering, delegation, skills, hooks, bootstrap.

中文简介正在准备中。当前显示原文说明。

hybridlabor-api/aos62026年10月16日 更新

agenttrail: live map of a multi-agent build in the browser: which plan component is being worked on, by which agent or harness, what is done and what is stuck. Use when a multi-agent pipeline starts (/startcycle, /startcycle-graph, /teamwork-preview) or after a plan-canvas approve, when the user asks to see what the agents are doing ("live map", "build board", "who is running now"), or by yourself whenever a multi-agent run is under way — no user prompt needed.

中文简介正在准备中。当前显示原文说明。

hybridlabor-api/aos62026年10月16日 更新

查看 hybridlabor-api 的全部技能

报告此技能的问题