Tier 1/2/3 system for designing agent skills.
-
Updated
May 14, 2026
Tier 1/2/3 system for designing agent skills.
A complete, open-source guide to Agent Skill design: from cognition to production. 28 chapters + 6 appendices.
A Skill design kernel for creating, auditing, rewriting, and evolving reusable AI Skills.
Turn real Agent failures into verifiable Skill boundaries and correct-layer fixes.
English translation of Vlad Yasko's claude-skill-module-design. 9-step engineering design methodology as a Claude Code skill.
A dual-loop self-evolution system for AI agents: fast fixes for local repeated mistakes, slow promotion for durable capability.
12 evidence-backed skill-writing principles plus repository engineering for real-world agent skills, inspired by Matt Pocock / 十二条有证据支撑的 Skill 写作原则与仓库工程指南
Engineer production-ready Codex skills with behavior contracts, quality lint, and forward testing / 将 Skill 构建从 vibe prompting 变成工程化流程
AI Product Teardown Workbench for classifying repositories, mapping AI surfaces, extracting reusable patterns, and drafting originality-safe new skills.
Add a description, image, and links to the skill-design topic page so that developers can more easily learn about it.
To associate your repository with the skill-design topic, visit your repo's landing page and select "manage topics."