meta-skill

Distilly — Distill how they think into reusable Skills for any Agent or Bot. Formerly Colleague Skill（原同事 Skill）.

YAO = Yielding AI Outcomes. A rigorous engineering, evaluation, governance, and portability system for reusable agent skills.

The meta-skill that builds and improves all your skills, including itself. Watches your work sessions (autonomous or human-led), captures patterns, corrections and judgement calls, and turns them into skill improvements and new skill candidates for your review. Practical application of the Augmented Expertise methodology. Open source: CC BY 4.0.

Making ANY Software Skill-Native -- Auto-generate production-ready AI Agent Skills for Claude Code, OpenClaw, Codex, and more.

Scholar-Wendao (学者问道) — open-source Claude Skill framework for distilling scholar perspectives into runnable analytical lenses. Generate perspective skills for humanities scholars (Stiegler, Aristotle, Foucault, …) as research lenses, not chat personas. Bilingual (EN/中文), supports contemporary + traditional scholars. By Shen Cong.

Health check for Claude Code skills. skill-checkup diagnoses why a SKILL.md won't reliably trigger, walks a dry-run of its workflow, and restructures the package. 4 deterministic scripts, no runtime deps.

A fork of Anthropic's skill-creator that fixes what the original got wrong: scripts Claude couldn't see, thin trigger tests, and a grader with no path to the tests it was supposed to grade.

Pro-grade lifecycle for creating, testing, versioning, and iterating on Claude Code skills — fuses Anthropic's skill-creator, lovstudio's pro-skills, and superpowers' writing-skills.

🔍 Discover and install AI skills for Claude Code. Automated search for behavioral frameworks, prompt-based skills, and specialized personas from GitHub with built-in security scanning.

Meta Skill Operating System — manages, analyzes, rates, optimizes, and generates Claude Code skills

Meta-skill that disciplines AI coding agents (Trae/Codex/Claude/Cursor/Windsurf) with 43 auditable work rules: risk grading (L1/L2/L3), rollback checkpoints, context self-heal, secret scanning, cross-platform adaptation. Constraints the agent itself, not its capabilities.

Research-first meta-skill that creates other Claude Code skills to an industry-leading, tested-real bar: scaffold to Skill-Repo-Spec, gate on evals/budget/dedup, then iterate via self-evolve. Thin orchestrator delegating research to market-intel and iteration to self-evolve.

A minimal meta-skill repository for building, refactoring, maintaining, extending, and recovering long-lived Hanako/OpenHanako skills.

AI coding chat portrait · 分析你跟 Claude Code / OpenCode / Kiro / Trae 的聊天记录,生成可视化性格画像。Self-reflection meta skill with conversation analytics, bilingual (zh/en), 100% local.

Agent team & skill factory for Claude Code — a maintained derivative of revfactory/harness, shipping its unmerged v2 rebuild plus reviewed community PRs, with CI that proves its own gates

Meta-skill: turns N papers + domain name into an A-grade queryable domain knowledge base (DAG + mental models).

zao-skill: AI meta-skill framework - design, verify, evolve production-grade agent skills

A Playground for AI Agents to learn DSP skills through trial and error

skill authoring framework

A triage-gated process for taking a behavioral skill from works to publish-grade and beyond. Scale the apparatus to the skill; every step leaves a checkable artifact.