By rfxlamia
Create and rigorously validate professional AI skills and subagents using dual-mode structured workflows—fast 12-step or full 15-step—including plan checks against DRY/YAGNI/TDD, behavioral pressure testing, anti-hallucination gates, and 7 production safety verifications with evidence collection for deployment.
npx claudepluginhub rfxlamia/skillkit --plugin skillkit-coreResearch-driven toolkit for creating and validating skills and subagents with structured workflows, gates, and automation scripts.
Validate implementation plans against DRY, YAGNI, TDD principles before execution.
Enforce 7 production safety gates with evidence before deployment.
The toolkit for creating reusable skills for AI coding agents.
SkillKit helps you build, validate, and share skills that extend how your AI agent works. Works with Claude Code, GitHub Copilot, OpenCode, and Codex.
Previously known as
claude-skillkit. All old links redirect here automatically.
SkillKit is a toolkit for creating and validating custom AI agent skills. It ships two core skills:
skillkit — meta-skill for creating new skills and subagents with dual-mode validationskillkit-help — pre-build orientation: understand what skills are, decide skills vs subagentsLooking for community skills & agents? Use:
npx @rfxlamia/sm→ SkillKit Marketplace
npx @rfxlamia/skillkit
Interactive CLI — pick your tool(s), choose user or project scope. Done.
| Skill | Trigger | What it does |
|---|---|---|
| skillkit | /skillkit | Create & validate skills/subagents — fast mode (12 steps) or full mode (15 steps) |
| skillkit-help | /skillkit-help | Answer "what are skills?", "should I make one?", "is my skill good enough?" |
npx @rfxlamia/skillkit/skillkit-help — guided path to your first skill in ~10 minutesThe skillkit skill drives the creation workflow:
| Trigger | Workflow | Steps |
|---|---|---|
create skill | Full skill creation | 12 steps with research + validation |
create subagent | Subagent creation | 8 steps with template-based workflow |
validate skill | Validation only | Structure, references, quality checks |
Skills vs Subagents | Decision helper | Recommends approach, then creates |
convert doc to skill | Migration | Transform existing docs into skills |
Quality target: 9.0+/10 via 5-layer validation and multi-proposal generation.
14 Python scripts in skills/skillkit/scripts/ — all support --format {text|json}.
| Script | Purpose |
|---|---|
init_skill.py | Initialize new skill directory structure |
init_subagent.py | Initialize new subagent with YAML template |
validate_skill.py | Structure and YAML validation |
token_estimator.py | Token consumption and cost estimation |
split_skill.py | Progressive disclosure auto-splitting |
pattern_detector.py | Workflow pattern recommendation |
pattern_detector_new.py | Enhanced pattern detection |
decision_helper.py | Skills vs Subagents decision tree |
security_scanner.py | Security vulnerability detection |
test_generator.py | Automated test generation |
quality_scorer.py | 5-category quality scoring (100 points) |
migration_helper.py | Document to skill conversion |
package_skill.py | Package skill for distribution |
quick_validate.py | Quick validation checks |
CLI installer

Guided orientation

A finished skill

24 community skills and 7 agents are available at SkillKit Marketplace:
npx @rfxlamia/sm
Apache 2.0 - Copyright 2025 rfxlamia
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
No model invocation
Executes directly as bash, bypassing the AI model
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
Comprehensive PR review agents specializing in comments, tests, error handling, type design, code quality, and code simplification
The most comprehensive Claude Code plugin — 38 agents, 156 skills, 72 legacy command shims, selective install profiles, and production-ready hooks for TDD, security scanning, code review, and continuous learning
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
Browser automation and end-to-end testing MCP server by Microsoft. Enables Claude to interact with web pages, take screenshots, fill forms, click elements, and perform automated browser testing workflows.