Extract
Initialize or refresh a gauntlet knowledge base from AST extraction plus AI-enriched explanations for codebase challenge workflows.
Overview
extract is an agent skill most often used in Build (also Ship review prep and Operate iterate) that compiles AST-derived codebase knowledge into .gauntlet/knowledge.json with AI-enriched gauntlet entries.
Install
npx skills add https://github.com/athola/claude-night-market --skill extractWhat is this skill?
- Runs python3 extractor.py on target directory for AST-based entry capture
- AI enrichment expands each entry detail with business logic, data flow, architecture, and rationale
- Cross-references modules via imports, shared types, and data-flow paths
- Merges with curated .gauntlet/annotations/ without overwriting hand-tuned entries
- Reports coverage gaps and difficulty distribution across 7 weighted categories
- 7 knowledge categories with explicit priority weights (business_logic weight 7 through error_handling weight 1)
Adoption & trust: 1 installs on skills.sh; 304 GitHub stars; 3/3 security scanners passed (skills.sh audits); trending (+100% hot-view momentum).
What problem does it solve?
Your gauntlet or agent challenges lack a structured, cross-linked map of business logic, APIs, and architecture in the repo.
Who is it for?
Developers maintaining athola/claude-night-market gauntlets who need reproducible codebase intelligence from AST plus narrative enrichment.
Skip if: Quick ad-hoc code search without gauntlet artifacts, or repos where running a Python extractor against the tree is disallowed.
When should I use this skill?
Initializing or refreshing codebase knowledge for gauntlet challenges—when .gauntlet/knowledge.json must be built or updated.
What do I get? / Deliverables
A merged knowledge.json with enriched details, annotation preservation, and a category or coverage summary ready for gauntlet challenge flows.
- .gauntlet/knowledge.json
- Coverage and difficulty summary report
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
Knowledge extraction supports builders who wire agent challenges to real repos—canonical shelf is build/agent-tooling because output feeds downstream gauntlet automation. Agent-tooling covers procedural skills that materialize structured repo intelligence (.gauntlet/knowledge.json) for agents, not app feature code.
Where it fits
First-time gauntlet setup on a monorepo by pointing extractor.py at the service root.
Regenerate knowledge after API contract changes so review challenges reference current endpoints.
Post-incident refresh to capture new error_handling paths in weighted categories.
How it compares
Structured gauntlet knowledge pipeline with weighted categories—not a generic codebase RAG indexer or static doc generator alone.
Common Questions / FAQ
Who is extract for?
Solo and indie builders using Claude Night Market gauntlets who want agent-ready codebase knowledge extracted from real project trees.
When should I use extract?
Use when initializing .gauntlet/knowledge.json, after major refactors in build, before ship review drills that need architecture context, or during operate iterate when refreshing challenge coverage.
Is extract safe to install?
It runs local Python extraction over your source tree; review the Security Audits panel on this page and audit scripts/extractor.py before execution.
SKILL.md
READMESKILL.md - Extract
# 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)