module-audit-agent
Use when asked to audit a module, check its health, or assess whether it should be refactored. Runs structural query, risk check, and test generatio…
安装
npx skills add wednesday-solutions/ai-agent-skills
收录说明
来自 wednesday-solutions/ai-agent-skills(172 stars),仓库最近更新 2026-05-26。尚未人工实测。
SKILL.md 原文
# Module Audit Agent
## When to use
- "Audit this module"
- "Is it safe to refactor X?"
- "What is the health of this service?"
- "Should we rewrite X?"
## What to do
1. **In parallel:**
- **brownfield-chat** — Read `dep-graph.json` and `summaries.json` for the target module. Report: what it does, its imports, what imports it, and any known conflicts.
- **brownfield-fix** — Run `wednesday-skills score <module>` and `wednesday-skills blast <module>`. Report risk band and total dependent count.
2. Present the combined audit report:
- Purpose summary (from summaries.json)
- Risk score + band (0–30 low / 31–60 medium / 61–80 high / 81–100 critical)
- Blast radius (dependent count, cross-language flagged separately)
- Architecture violations or danger zone warnings (from MASTER.md)
- Recommendation: proceed / review / senior sign-off / do not touch
3. **brownfield-tests** — Only run if coverage < 30% AND risk > 50:
- Run `wednesday-skills gen-tests --file <module>`
- Show generated test file to dev for review before writing
## Never
- Recommend refactoring a critical file (risk > 80) without flagging the blast radius
- Read raw source to answer structural questions — use graph only
- Auto-write test files without showing the dev first
同类技能
- api-and-interface-designGuides stable API and interface design. Use when designing APIs, module boundaries, or any public interface.…
- design-taste-frontendtasteskill v2, the anti-slop frontend skill. Art direction for websites, distilled from a study of 247 curat…
- understand-chatUse when you need to ask questions about a codebase or understand code using a knowledge graph
- artifacts-builderSuite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web te…
- open-code-review-delegateDelegation mode for open-code-review (OCR). Instead of OCR calling an LLM endpoint, this skill instructs the…
- cognee-cliUse when the user wants to drive cognee from the terminal with cognee-cli — remember/recall/forget/improve m…