# Extract

> Build a codebase knowledge base of business logic, architecture, data flow, and patterns.

- Skill: `majiayu000/extract-3` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add majiayu000/extract-3`
- Raw SKILL.md: https://api.skillmd.com/api/skills/majiayu000/extract-3/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Research & Search
- Author: majiayu000 (https://skillmd.com/u/majiayu000)
- Updated: 2026-09-09
- Page: https://skillmd.com/skills/majiayu000/extract-3

---


# Extract Codebase Knowledge

Build or rebuild the `.gauntlet/knowledge.json` knowledge base.

## Steps

1. **Identify target directory**: use the current working directory
   or a user-specified path

2. **Run AST extraction**: invoke the extractor script
   ```bash
   python3 ${CLAUDE_PLUGIN_ROOT}/scripts/extractor.py <target-dir>
   ```

3. **AI enrichment**: for each extracted entry, enhance the `detail`
   field with natural language explanation of business logic, data
   flow, architectural role, and rationale

4. **Cross-reference**: link related entries across modules by
   matching imports, shared types, and data flow paths

5. **Merge with annotations**: preserve existing curated entries
   in `.gauntlet/annotations/`

6. **Save**: write to `.gauntlet/knowledge.json`

7. **Report**: show summary by category, coverage gaps, difficulty
   distribution

## Category Priority

1. business_logic (weight 7)
2. architecture (weight 6)
3. data_flow (weight 5)
4. api_contract (weight 4)
5. pattern (weight 3)
6. dependency (weight 2)
7. error_handling (weight 1)