Professional skill creation with research-driven workflow and automated validation. USE WHEN: Creating new skills, validating existing skills, deciding between Skills vs Subagents, migrating documents to skills, or running individual validation tools. PRIMARY TRIGGERS: "create skill" = Full creation (12 steps with research + execution planning) "validate skill" = Validation workflow (steps 3-8) "Skills vs Subagents" = Decision workflow (step 0) "convert doc to skill" = Migration workflow "estimate tokens" = Token optimization "security scan" = Security audit WORKFLOW COMPLIANCE: Structured workflows with validation checkpoints. Research phase (Step 1c-1d) ensures skills based on proven approaches. DIFFERENTIATOR: Research-driven creation. Web search (3-5 queries) before building. Multi-proposal generation. 9 automation scripts. Quality 9.0+/10. REUSED: Anthropic's init_skill.py and package_skill.py (production-tested).
View on GitHubrfxlamia/claude-skillkit
claude-skillkit
January 20, 2026
Select agents to install to:
npx add-skill https://github.com/rfxlamia/claude-skillkit/blob/main/skills/claude-skillkit/SKILL.md -a claude-code --skill claude-skillkitInstallation paths:
.claude/skills/claude-skillkit/## Section 1: Intent Detection & Routing **Detect user intent, route to appropriate workflow.** | Intent | Keywords | Route To | |--------|----------|----------| | Full creation | "create", "build", "new skill" | Section 2 | | Validation | "validate", "check quality" | Section 3 | | Decision | "Skills vs Subagents", "decide" | Section 4 | | Migration | "convert", "migrate doc" | Section 5 | | Single tool | "validate only", "estimate tokens", "scan" | Section 6 | **PROCEED to corresponding section after intent detection.** **Workflow Value:** Research-driven approach validates design before building. Sequential steps with checkpoints produce 9.0/10+ quality vs ad-hoc creation. --- ## Section 2: Full Creation Workflow (Overview) **Prerequisites:** Skill description provided, workspace available **Quality Target:** >=7.5/10 (Good), >=8.0/10 (Excellent) - See v1.2.1 quality improvements **Time:** <10 min with automation ### 12-Step Process with Validation Gates: **STEP 0: Decide Approach** - Tool: `decision_helper.py` - Decides: Skills vs Subagents - Gate: Proceed only if "Skills" recommended **STEP 1: Understand & Research** - 1a. Gather requirements - 1b. Identify knowledge gaps - 1c. Research domain (Verbalized Sampling: 3-4 web searches with diverse angles) - 1d. Generate proposals (3-5 options evaluated with multi-criteria scoring) - 1e. User validates and approves approach - 1f. Execution planning: P0/P1/P2 prioritization with token budgets assigned - See: `references/section-2-full-creation-workflow.md` (Step 1f details) **STEP 2: Initialize & Create Content** - Tool: `python scripts/init_skill.py skill-name --path /path` (Anthropic) - Alternative: `migration_helper.py` (if converting from document) - 2.5 Checkpoint: Sequential creation (P0→P1→P2), token budget monitoring - 2.8 Verification: P0/P1/P2 completion validation before proceeding - See: `references/section-2-full-creation-workflow.md` (Steps 2.5 & 2.8 details) **STEP 3: Validate Structure