setup-matt-pocock-skills
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChineseSetup Matt Pocock's Skills
配置Matt Pocock的Skills
Scaffold the per-repo configuration that the engineering skills assume:
- Backlog — where issues/tickets live (GitHub by default; local markdown is also supported out of the box)
- Triage labels — the strings used for the five canonical triage roles
- Domain docs — where and ADRs live, and the consumer rules for reading them
CONTEXT.md
This is a prompt-driven skill, not a deterministic script. Explore, present what you found, confirm with the user, then write.
搭建工程技能所需的仓库专属配置:
- 待办事项(Backlog) — 问题/工单的存储位置(默认是GitHub;也原生支持本地Markdown)
- 分类标签(Triage labels) — 五个标准分类角色所使用的字符串
- 领域文档(Domain docs) — 和架构决策记录(ADRs)的存储位置,以及读取这些文档的规则
CONTEXT.md
这是一个由提示驱动的技能,而非确定性脚本。流程为:探查现状、展示发现、与用户确认、然后写入配置。
Process
流程
1. Explore
1. 探查
Look at the current repo to understand its starting state. Read whatever exists; don't assume:
- and
git remote -v— is this a GitHub repo? Which one?.git/config - and
AGENTS.mdat the repo root — does either exist? Is there already anCLAUDE.mdsection in either?## Agent skills - and
CONTEXT.mdat the repo rootCONTEXT-MAP.md - and any
docs/adr/directoriessrc/*/docs/adr/ - — does this skill's prior output already exist?
docs/agents/ - — sign that a local-markdown backlog convention is already in use
.scratch/
查看当前仓库以了解初始状态。读取所有已存在的内容,不要做假设:
- 和
git remote -v— 这是GitHub仓库吗?具体是哪个?.git/config - 仓库根目录下的和
AGENTS.md— 二者是否存在?其中是否已有CLAUDE.md章节?## Agent skills - 仓库根目录下的和
CONTEXT.mdCONTEXT-MAP.md - 以及任何
docs/adr/目录src/*/docs/adr/ - — 本技能之前的输出是否已存在?
docs/agents/ - — 表示已在使用本地Markdown待办事项约定的标志
.scratch/
2. Present findings and ask
2. 展示发现并询问
Summarise what's present and what's missing. Then walk the user through the three decisions one at a time — present a section, get the user's answer, then move to the next. Don't dump all three at once.
Assume the user does not know what these terms mean. Each section starts with a short explainer (what it is, why these skills need it, what changes if they pick differently). Then show the choices and the default.
Section A — Backlog backend.
Explainer: "Backlog" is where issues/tickets live for this repo. Skills like,to-issues,triage, andto-prdread from and write to it — they need to know whether to callqa, write a markdown file undergh issue create, or follow some other workflow you describe. Pick the place you actually track work for this repo..scratch/
Default posture: these skills were designed for GitHub. If a points at GitHub, propose that. Otherwise (or if the user prefers), offer:
git remote- GitHub — issues live in the repo's GitHub Issues
- Local markdown — issues live as files under in this repo (good for solo projects or repos without a GitHub remote)
.scratch/<feature>/ - Other (Jira, Linear, etc.) — ask the user to describe the workflow in one paragraph; the skill will record it as freeform prose
Section B — Triage label vocabulary.
Explainer: When theskill processes an incoming issue, it moves it through a state machine — needs evaluation, waiting on reporter, ready for an AFK agent to pick up, ready for a human, or won't fix. To do that, it needs to apply labels (or the equivalent in your backlog) that match strings you've actually configured. If your repo already uses different label names (e.g.triageinstead ofbug:triage), map them here so the skill applies the right ones instead of creating duplicates.needs-triage
The five canonical roles:
- — maintainer needs to evaluate
needs-triage - — waiting on reporter
needs-info - — fully specified, AFK-ready (an agent can pick it up with no human context)
ready-for-agent - — needs human implementation
ready-for-human - — will not be actioned
wontfix
Default: each role's string equals its name. Ask the user if they want to override any. If their backlog has no existing labels, the defaults are fine.
Section C — Domain docs.
Explainer: Some skills (,improve-codebase-architecture,diagnose) read atddfile to learn the project's domain language, andCONTEXT.mdfor past architectural decisions. They need to know whether the repo has one global context or multiple (e.g. a monorepo with separate frontend/backend contexts) so they look in the right place.docs/adr/
Confirm the layout:
- Single-context — one +
CONTEXT.mdat the repo root. Most repos are this.docs/adr/ - Multi-context — at the root pointing to per-context
CONTEXT-MAP.mdfiles (typically a monorepo).CONTEXT.md
总结已存在的内容和缺失的部分。然后逐一引导用户完成三个决策 — 展示一个环节、获取用户的答案、再进入下一个环节。不要一次性抛出所有三个问题。
假设用户不了解这些术语的含义。每个环节都以简短的解释开头(是什么、这些技能为何需要它、选择不同选项会有什么变化)。然后展示选项和默认值。
环节A — 待办事项后端
解释:“待办事项(Backlog)”是本仓库中问题/工单的存储位置。、to-issues、triage和to-prd等技能会从中读取并写入内容 — 它们需要知道是调用qa、在gh issue create下写入Markdown文件,还是遵循你描述的其他工作流程。请选择你实际用于跟踪本仓库工作的位置。.scratch/
默认设置:这些技能是为GitHub设计的。如果指向GitHub,则建议使用该选项。否则(或用户偏好其他选项),提供以下选择:
git remote- GitHub — 问题存储在仓库的GitHub Issues中
- 本地Markdown — 问题作为文件存储在本仓库的下(适用于个人项目或没有GitHub远程仓库的项目)
.scratch/<feature>/ - 其他(Jira、Linear等) — 请用户用一段话描述工作流程;本技能会将其作为自由格式文本记录下来
环节B — 分类标签词汇
解释:当技能处理新问题时,它会通过状态机推进问题:需要评估、等待报告人回复、准备好由AFK agent接手、准备好由人工处理、或不会修复。要实现这一点,它需要应用与你实际配置的字符串匹配的标签(或待办事项系统中的等效标记)。如果你的仓库已经使用不同的标签名称(例如triage而非bug:triage),请在此处映射它们,以便技能应用正确的标签而非创建重复标签。needs-triage
五个标准角色:
- — 需要维护者评估
needs-triage - — 等待报告人回复
needs-info - — 已完全明确,可由AFK agent接手(无需人工上下文即可处理)
ready-for-agent - — 需要人工实现
ready-for-human - — 不会采取行动
wontfix
默认值:每个角色的字符串与其名称一致。询问用户是否需要覆盖任何默认值。如果你的待办事项系统没有现有标签,默认值即可。
环节C — 领域文档
解释:部分技能(、improve-codebase-architecture、diagnose)会读取tdd文件以了解项目的领域语言,并读取CONTEXT.md中的过往架构决策记录。它们需要知道仓库是采用单一全局上下文还是多上下文(例如包含独立前端/后端上下文的单体仓库),以便在正确的位置查找文档。docs/adr/
确认布局:
- 单上下文 — 仓库根目录下有一个+
CONTEXT.md。大多数仓库属于此类。docs/adr/ - 多上下文 — 根目录下的指向各个上下文对应的
CONTEXT-MAP.md文件(通常是单体仓库)。CONTEXT.md
3. Confirm and edit
3. 确认并编辑
Show the user a draft of:
- The block to add to whichever of
## Agent skills/CLAUDE.mdis being edited (see step 4 for selection rules)AGENTS.md - The contents of ,
docs/agents/backlog.md,docs/agents/triage-labels.mddocs/agents/domain.md
Let them edit before writing.
向用户展示以下内容的草稿:
- 要添加到/
CLAUDE.md中任一文件的AGENTS.md块(选择规则见步骤4)## Agent skills - 、
docs/agents/backlog.md、docs/agents/triage-labels.md的内容docs/agents/domain.md
让用户在写入前进行编辑。
4. Write
4. 写入
Pick the file to edit:
- If exists, edit it.
CLAUDE.md - Else if exists, edit it.
AGENTS.md - If neither exists, ask the user which one to create — don't pick for them.
Never create when already exists (or vice versa) — always edit the one that's already there.
AGENTS.mdCLAUDE.mdIf an block already exists in the chosen file, update its contents in-place rather than appending a duplicate. Don't overwrite user edits to the surrounding sections.
## Agent skillsThe block:
markdown
undefined选择要编辑的文件:
- 如果存在,则编辑它。
CLAUDE.md - 否则如果存在,则编辑它。
AGENTS.md - 如果二者都不存在,请询问用户要创建哪一个 — 不要自行选择。
当已存在时,切勿创建(反之亦然) — 始终编辑已存在的文件。
CLAUDE.mdAGENTS.md如果所选文件中已存在块,请原地更新其内容,而非追加重复块。不要覆盖用户对周围章节的编辑。
## Agent skills块内容如下:
markdown
undefinedAgent skills
Agent skills
Backlog
Backlog
[one-line summary of where the backlog lives]. See .
docs/agents/backlog.md[待办事项存储位置的一行摘要]。详见。
docs/agents/backlog.mdTriage labels
Triage labels
[one-line summary of the label vocabulary]. See .
docs/agents/triage-labels.md[标签词汇的一行摘要]。详见。
docs/agents/triage-labels.mdDomain docs
Domain docs
[one-line summary of layout — "single-context" or "multi-context"]. See .
docs/agents/domain.md
Then write the three docs files using the seed templates in this skill folder as a starting point:
- [backlog-github.md](./backlog-github.md) — GitHub backlog
- [backlog-local.md](./backlog-local.md) — local-markdown backlog
- [triage-labels.md](./triage-labels.md) — label mapping
- [domain.md](./domain.md) — domain doc consumer rules + layout
For "other" backlog backends, write `docs/agents/backlog.md` from scratch using the user's description.[布局的一行摘要 — “单上下文”或“多上下文”]。详见。
docs/agents/domain.md
然后以本技能文件夹中的种子模板为起点,写入三个文档文件:
- [backlog-github.md](./backlog-github.md) — GitHub待办事项
- [backlog-local.md](./backlog-local.md) — 本地Markdown待办事项
- [triage-labels.md](./triage-labels.md) — 标签映射
- [domain.md](./domain.md) — 领域文档读取规则 + 布局
对于“其他”待办事项后端,请根据用户的描述从头编写`docs/agents/backlog.md`。5. Done
5. 完成
Tell the user the setup is complete and which engineering skills will now read from these files. Mention they can edit directly later — re-running this skill is only necessary if they want to switch backlog backends or restart from scratch.
docs/agents/*.md告知用户配置已完成,以及哪些工程技能将从此读取文件。提醒用户之后可以直接编辑 — 只有当他们想要切换待办事项后端或从头开始重新配置时,才需要重新运行本技能。
docs/agents/*.md