liaosw97/skill-craft-adapter
Claude Code Skill 质量评估插件——基于加权评分框架（8 模块检查、7 反模式评估、3 完备性原则、决策门），用于评估、修复、创建和审计 Claude Code Skill。
Claude Code Skill 质量评估插件——基于加权评分框架（8 模块检查、7 反模式评估、3 完备性原则、决策门），用于评估、修复、创建和审计 Claude Code Skill。
npx skills add liaosw97/skill-craft-adapterClaude Code Skill 质量评估插件——基于加权评分框架（8 模块检查、7 反模式评估、3 完备性原则、决策门），用于评估、修复、创建和审计 Claude Code Skill。
CODEX SKILL regarding short-term and long-term memory management
Lightweight Long-Term Memory for LLM Agents.
One-shot 'upgrade anything' skill for Claude Code: leads with new features, writes the brief, plans, builds with subagents, reviews via an expert panel + codex, improves until world-class, and self-learns. Works on code, UI, copy, and strategy. Self-contained with built-in fallbacks.
Fetch any URL and convert to markdown using baoyu-fetch CLI (Chrome CDP with site-specific adapters). Built-in adapters for X/Twitter, YouTube transcripts, Hacker News threads, and generic pages via Defuddle. Handles login/CAPTCHA via interaction wait modes. Use when user wants to save a webpage as markdown.
Ten Commandments for Coding Agents — a minimal AGENTS.md / CLAUDE.md drop-in for Claude Code, Codex, Gemini CLI, OpenCode, Cursor and any tool that reads agent rules. KO/EN bilingual. GitHub Pages: https://cskwork.github.io/coding-agent-rules/
Debian packaging skill with comprehensive policy, debhelper, and language-specific knowledge for Ruby, Python, Rust, and Go. This is largely LLM-generated, thoughtfully. PRs to tighten this and remove slop appreciated. I mostly use this for personal, private packages not OSS Debian packages. This skill isn't ready for that.