to-issues
vinvcn/mattpocock-skills-zh-cn
Break plans into independent vertical-slice issues for your tracker using tracer-bullet decomposition.
What is to-issues?
Decomposes specifications, PRDs, or plans into thin vertical slices that can be independently assigned and completed in an issue tracker. Use this when you need to convert a plan into actionable issues, create implementation tasks, or break work into granular, end-to-end deliverables.
- Analyzes plans, specs, or PRDs and identifies vertical slices (tracer bullets) that span all integration layers
- Distinguishes between HITL (human-in-the-loop) and AFK (autonomous) slices to prioritize work
- Generates issue tracker-ready issues with acceptance criteria, dependencies, and user story mappings
- Explores codebase context to ensure issues use project domain vocabulary and follow ADRs
- Iterates with you on granularity and dependency relationships before publishing
How to install to-issues
npx skills add https://github.com/vinvcn/mattpocock-skills-zh-cn --skill to-issues- Issue tracker already set up with triage label vocabulary
- Optionally run `/setup-matt-pocock-skills` to configure issue tracker access
- Access to the codebase (for optional exploration to understand current state)
How to use to-issues
- 1.Provide the skill with a plan, specification, PRD, or issue reference (number, URL, or path)
- 2.The skill gathers context from conversation and optionally explores the codebase
- 3.Review the proposed vertical-slice breakdown presented as a numbered list with titles, types (HITL/AFK), blockers, and user stories
- 4.Provide feedback on granularity, dependencies, and slice categorization; iterate until approved
- 5.The skill publishes approved issues to your tracker in dependency order, using the standard issue template
Use cases
- Converting a feature specification into a backlog of independently assignable issues
- Breaking down a large PRD into thin vertical slices that deliver end-to-end value
- Creating implementation tasks that span schema, API, UI, and tests in a single slice
- Identifying which work requires human decisions (HITL) versus what can be automated (AFK)
- Establishing correct dependency order for issues so blockers are completed first
- Engineering managers planning sprints or releases
- Developers breaking down features into implementable tasks
- Tech leads designing work decomposition for teams
- Product managers converting specs into actionable issues
to-issues FAQ
A thin, end-to-end path through all integration layers (schema, API, UI, tests) that delivers a complete, demonstrable piece of functionality—not a horizontal slice of just one layer.
HITL (human-in-the-loop) slices require human decisions like architecture reviews or design approvals. AFK (away from keyboard) slices can be implemented and merged autonomously. The skill prioritizes AFK when possible.
Yes, your issue tracker and triage label vocabulary should be configured. Run `/setup-matt-pocock-skills` if needed to establish the connection.
Yes. The skill presents the breakdown and asks for feedback on granularity, dependencies, and categorization. You can request merging, splitting, or reordering slices before publishing.
No. The skill only publishes new issues from the approved breakdown. It does not close or modify any parent issues.
Full instructions (SKILL.md)
Source of truth, from vinvcn/mattpocock-skills-zh-cn.
name: to-issues description: 使用 tracer-bullet 垂直切片,把计划、规格或 PRD 拆成项目 issue tracker 上可独立领取的 issues。适用于用户想把计划转成 issues、创建实现议题,或把工作拆成 issues 时。
To Issues
使用 vertical slices(tracer bullets)把计划拆成可独立领取的 issue tracker issues。
Issue tracker 和 triage label vocabulary 应该已经提供给你;如果没有,运行 /setup-matt-pocock-skills。
Process
1. Gather context
基于 conversation context 中已有内容工作。如果用户传入 issue reference(issue number、URL 或 path)作为参数,从 issue tracker 获取它,并读取完整 body 和 comments。
2. Explore the codebase (optional)
如果还没探索 codebase,就先探索,以理解代码当前状态。Issue titles 和 descriptions 应使用项目 domain glossary vocabulary,并遵守相关 ADRs。
3. Draft vertical slices
把计划拆成 tracer bullet issues。每个 issue 都是一个薄 vertical slice,end-to-end 穿过所有 integration layers,而不是某一层的 horizontal slice。
Slices 可以是 HITL 或 AFK。HITL slices 需要人类交互,例如 architecture decision 或 design review。AFK slices 可以无人交互地实现并合并。尽可能优先 AFK。
4. Quiz the user
把 proposed breakdown 作为编号列表展示。每个 slice 显示:
- Title:短描述名
- Type:HITL / AFK
- Blocked by:哪些其他 slices 必须先完成(如果有)
- User stories covered:覆盖哪些 user stories(如果 source material 中有)
询问用户:
- Granularity 是否合适?(too coarse / too fine)
- Dependency relationships 是否正确?
- 是否需要 merge 或继续 split 某些 slices?
- HITL 和 AFK 标记是否正确?
迭代直到用户批准 breakdown。
5. Publish the issues to the issue tracker
对每个批准的 slice,把新 issue 发布到 issue tracker。使用下面的 issue body template。这些 issues 被视为已准备好给 AFK agents 接手;除非另有指示,发布时应用正确的 triage label。
按 dependency order 发布 issues(blockers first),这样可以在 "Blocked by" 字段引用真实 issue identifiers。
<issue-template> ## Parent对 issue tracker 中 parent issue 的引用(如果 source 是现有 issue;否则省略本 section)。
What to build
这个 vertical slice 的简洁描述。描述 end-to-end behavior,不要按 layer-by-layer implementation 描述。
避免具体 file paths 或 code snippets;它们很快会过时。例外:如果 prototype 产出的 snippet 比 prose 更精确地编码了某个决策(state machine、reducer、schema、type shape),可以内联在这里,并简短说明它来自 prototype。保留决策密集部分,不要放完整 working demo。
Acceptance criteria
- Criterion 1
- Criterion 2
- Criterion 3
Blocked by
- 对 blocking issue 的引用(如果有)
如果没有 blocker,写 "None - can start immediately"。
</issue-template>不要 close 或 modify 任何 parent issue。
Related skills
More from vinvcn/mattpocock-skills-zh-cn and the wider catalog.

to-prd
Convert conversation context into a structured PRD and publish to your project's issue tracker.

to-questionnaire
Convert decisions you can't answer alone into a structured questionnaire for others to complete.

to-spec
Convert conversation context into a spec and publish to your project issue tracker.

to-tickets
Break plans, specs, or conversations into tracer-bullet tickets with blocking dependencies for your issue tracker.

triage
Triage issues and PRs through a state machine of roles, verify claims, and generate agent-ready briefs.

ubiquitous-language
Extract and formalize domain terminology from conversations into a DDD-style ubiquitous language glossary.