xxih/ai-harness-zh/investigate
Trust tier 3
- community
Source check pending
Checking that the SKILL.md file is still in the repository.
系统化 root-cause debugging。遵守“没有调查就没有修复”的铁律， 先锁范围、看模式、测假设，再实现修复与验证。 (gstack)
Install
npx vskill@latest install xxih/ai-harness-zh/investigateSKILL.md
/investigate
/investigate 是 gstack 的排障铁律：没有调查，就没有修复。
Iron Law
- 不允许先拍脑袋改代码，再事后编造原因。
- 必须先定位根因，再进入实现。
核心流程
Phase 1: Root Cause Investigation
- 复现问题
- 追踪数据流 / 调用链
- 读日志、错误、相关文件
Prior Learnings
- 查历史 learnings，看是不是重复问题
Scope Lock
- 把调查范围锁到相关模块，避免“顺手修别处”
Phase 2: Pattern Analysis
- 看这是个单点 bug、系统性模式、还是已知架构味道
Phase 3: Hypothesis Testing
- 列假设
- 一个个证伪 / 证实
Phase 4: Implementation
- 只有在根因足够明确后才开始修
Phase 5: Verification & Report
- 证明修复有效，写清证据
Capture Learnings
- 把这次调查的可复用洞察沉淀下来
关键规则
- 连续 3 次失败修复后要停下来，重新诊断，不许继续乱试。
- 没有验证证据，就不能宣布完成。
Security
- Security scanPassed, 95 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingreferences/translations/gstack/investigate/SKILL.md
Popularity
- Repository stars
- 1
- 7-day trend score
- 8
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/xxih/ai-harness-zh/investigate)