# skill-extractor

skill-extractor turns a finished session into a reusable Claude skill: the cause of a failure, a workaround, the symptom that points to it.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf (Devin Desktop), Claude.ai · ★ 540 repo stars
- Author: Trail of Bits
- License: CC-BY-SA-4.0 · policy=link-only · description only; the file stays with the author
- Source: https://github.com/trailofbits/skills-curated/tree/HEAD/plugins/skill-extractor/skills/skill-extractor/SKILL.md
- Updated: 2026-07-14
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/skills/skill-extractor

## What it does

Point it at the session you just finished and it pulls out what deserves a second life: the cause behind a non-obvious failure, a workaround found by trial and error, the symptom that leads to either. Output is a `SKILL.md` file. It lands in `~/.claude/skills/[name]/SKILL.md` by default, or in `.claude/skills/[name]/SKILL.md` when you pass `--project`, which means it ships with the repo.

The process is ordered. First it looks through both skill directories with `ls` and `grep` for an existing skill that already covers the topic, because updating beats piling up duplicates. Then it names the problem, the insight and the trigger conditions, scores the candidate against a criteria table and asks you to confirm. A kebab-case name and a scope come next. Research through web search or Context7 MCP is optional and only for library or framework topics. Template, validation, save.

Nothing fires by itself: there are no hooks or automatic triggers, and `/skill-extractor` is the only way in. Tool access stops at `Read`, `Write`, `Glob`, `Grep`, `WebSearch` and `AskUserQuestion`. It also declines thin material — trivial fixes, one-off project config, anything already covered elsewhere, and fixes that have not been shown to work.

## Install

### Claude Code (native platform)

1. **Plugin marketplace**
In the agent chat (not a shell):
```text
/plugin marketplace add trailofbits/skills-curated
```
In the agent chat (not a shell):
```text
/plugin install skill-extractor@skills-curated
```
This installs the whole skill-extractor plugin with all its skills

2. **Skills CLI (Vercel)**
```bash
npx skills add trailofbits/skills-curated --skill skill-extractor -a claude-code
```

3. **Manually: put the files in place**
- File: `~/.claude/skills/skill-extractor/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/skill-extractor/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add trailofbits/skills-curated --skill skill-extractor -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/skill-extractor/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add trailofbits/skills-curated --skill skill-extractor -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/skill-extractor/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install trailofbits/skills-curated plugins/skill-extractor/skills/skill-extractor
```

2. **Manually: put the files in place**
- File: `.github/skills/skill-extractor/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add trailofbits/skills-curated --skill skill-extractor -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/trailofbits/skills-curated --path plugins/skill-extractor/skills/skill-extractor
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/skill-extractor/SKILL.md` — The legacy .windsurf/skills/ path still works

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/trailofbits/skills-curated/tree/HEAD/plugins/skill-extractor/skills/skill-extractor/SKILL.md

## When to use it
- capture a hard-won bug fix as a reusable skill
- document a workaround that took trial and error
- record the exact error text that should trigger the knowledge next time
- update an existing skill with a new edge case instead of adding a duplicate
- filter candidates through a quality checklist before saving

## Example prompts
- \`/skill-extractor` — turn the AST recursion crash we just fixed into a skill
- \`/skill-extractor --project` — keep this one in the repo next to the code
- \`/skill-extractor` — check whether a retry skill already exists and update that one

## Similar
- [wiki](https://affpapa.org/en/nevibecode/skills/yeachan-heo-wiki.md) — skill · 685 installs
- [Documentation Extractor](https://affpapa.org/en/nevibecode/mcp/documentation-extractor.md) — MCP server · ★ 35 repo stars
- [compile-knowledge](https://affpapa.org/en/nevibecode/skills/compile-knowledge.md) — skill · 3300 installs
- [openclaw-history-ingest](https://affpapa.org/en/nevibecode/skills/openclaw-history-ingest.md) — skill · 2200 installs
- [data-context-extractor](https://affpapa.org/en/nevibecode/skills/data-context-extractor.md) — skill · 3300 installs
- [story-memory](https://affpapa.org/en/nevibecode/skills/story-memory.md) — skill · 606 installs

From the same set: [trailofbits/skills-curated](https://affpapa.org/en/nevibecode/by/trailofbits/skills-curated.md) — 51 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
