explain-how
Explain current code with file references and explorer subagents. Use for unfamiliar code or how-it-works questions; use explain-why for history or rationale, or a bare explorer Task for quick lookups.
Cynnamoroll · 已有静态检查记录
安装与使用
运行命令前先阅读来源与依赖说明。复制命令不会自动安装或执行。
npx skills add Cynnamoroll/factory-stack --skill explain-how技能说明
name: explain-how description: "Explain current code with file references and explorer subagents. Use for unfamiliar code or how-it-works questions; use explain-why for history or rationale, or a bare explorer Task for quick lookups." license: MIT compatibility: droid allowed-tools:
- Read
- Grep
- Glob
- LS
- Task
Explain how
Give a senior engineer a working mental model, not annotated source. Stay read-only.
1. Size the question
If scope is unclear, state your reading and continue.
- Simple (one module, one function, a narrow question). Read the code yourself and go to step 3.
- Complex (a subsystem across files or services, a cross-cutting feature, an architecture overview). Go to step 2.
When in doubt, take the simple path.
2. Explore in parallel
Split the question into 2 to 4 angles, each a distinct slice of the subsystem. Issue one explorer Task per angle, all in one message. Build each prompt from references/explorer-prompt.md with the question and the angle filled in.
3. Write the walkthrough
Merge explorer reports and settle contradictions from code without re-exploring. Follow references/walkthrough-format.md, omitting inapplicable sections.
Done when
- Every claim about behavior cites a real
file:lineor symbol you or an explorer read. - Open questions the explorers could not trace are listed, not hidden.
- No file was changed.