Skip to content
🤖 Consolidated, AI-optimized SKF docs: llms-full.txt. Fetch this plain text file for complete context.

Skill Forge

Every instruction your AI reads cites a file, a line, and a commit SHA.

AI agents hallucinate API calls. They invent function names, guess parameter types, and produce code that doesn’t compile.

Skill Forge reads the source and hands your agent the truth — with receipts. Every function signature, every parameter type, every usage pattern traces back to a file, a line, and a commit SHA in the upstream repository.

A receipt looks like [AST:cognee/api/v1/search/search.py:L26]

If SKF can’t cite a source, it doesn’t include the instruction.

Verify any claim in 60 seconds →

ApproachWhat it does wellWhere it falls short
Skill scaffolding (npx skills init)Generates a spec-compliant skill fileThe file is empty — you still have to write every instruction by hand
LLM summarizationUnderstands context and intentGenerates plausible-sounding content that may not match the actual API
RAG / context stuffingRetrieves relevant code snippetsReturns fragments without synthesis — no coherent skill output
Manual authoringHigh initial accuracyDrifts as the source code changes, doesn’t scale across dependencies
IDE built-in context (Copilot, Cursor)Convenient, zero setupUses generic training data, not your project’s specific integration patterns
Skill ForgeEvery instruction cites upstream file:line at a pinned commit. Falsifiable in 60 seconds.Coverage depends on which tools you’ve installed (Quick / Forge / Forge+ / Deep tiers).

Requires Node.js >= 22, Python >= 3.10, and uv.

Terminal window
npx bmad-module-skill-forge install

Then generate your first skill:

@Ferris SF # Set up your forge
@Ferris QS <package> # Generate a skill in under a minute

See Getting Started for platform support, tier selection, and troubleshooting.