skill-crafter

name	skill-crafter
description	Build, improve, modularize, and validate Claude Code skills the right way. Use when creating a new skill, improving or auditing an existing one, refactoring a SKILL.md that grew too long, or adding validation scripts. Encodes a slim-router, progressive-disclosure, PACT-driven approach and is its own reference implementation.

Skill Crafter

Context: a meta-skill for building and improving other skills. It encodes a set of best practices and demonstrates every one of them in its own structure. You reason through the work; the scripts verify what is checkable. The nine principles every skill should satisfy live in references/best-practices.md.

Create a new skill

Follow protocols/create-skill.md, which runs the PACT cycle:

1. Prepare research the domain via agents/skill-preparer.md.
2. Architect design the tree and dependency map via agents/skill-architect.md.
3. Create build the files via agents/skill-creator.md.
4. Test validate via agents/skill-tester.md and protocols/validate.md.

Every skill it produces is wired to refine itself, using the self-refine and refinement-log templates.

Improve an existing skill

Point at the target skill and follow protocols/improve-skill.md: assess it against the practices with agents/skill-improver.md, apply the fixes, and re-test.

Single tasks

• protocols/modularize.md to split a SKILL.md that grew past a slim router.
• protocols/validate.md to check a skill against the rules.

Map

• references/ the why: best practices, progressive disclosure, folder taxonomy, prompt structure, validation pattern. Read on demand.
• protocols/ the how: create-skill, improve-skill, modularize, validate.
• agents/ the workers: PACT (preparer, architect, creator, tester) plus improver.
• scripts/ CLI tools: scripts/scaffold.py to generate a skill, scripts/validate_skill.py to check its structure and completeness, and scripts/verify_package.py to round-trip the packaged .skill against source. Run them, do not reimplement.
• templates/ files to copy into a new skill: SKILL, agent, protocol, reference, script, plus the self-refine protocol and refinement-log that make a skill self-refining.