用六步法拆解 Agent Skill,并生成精简、可复用的分析报告
Agents
Designing, evaluating, and shipping autonomous AI agents.
9632 posts
AI学术工作流 · 乔淼PhD「AI学术训练营」skills 库:选题 / 文献搜索 / 文献分析 / 研究设计 / 数据分析 / 全文初稿,30 个 skills 拆到步骤颗粒度,...
Human-in-the-loop skill for AI agents — create tasks for humans to review, approve, or complete. REST API + MCP.
Claude Code skill: turn a keyword into presentation-ready PNG icons — 3 concepts, 1:1, transparent, recolorable
Agent skill: recursive plan with mechanically gated leaves + dependency-safe subagent waves — install via npx skills
You are a different collaborator on every project. Your session history already knows. Reads your local Claude Code / Codex history and writ...
深知写作助手:基于深知搜索的公文与政务材料写作 Agent Skill
Enforce clear technical prose. Remove AI slop.
Skill for building memory by agent itself
让 deepseek V4 等没有 vision 能力的大模型也能识图——符合 Agent Skills 开放标准,跨 Claude Code/Codex/opencode/Cursor 等主流 agent 通用。
Drop in a picture — your cat, a doodle, a game character — and Claude analyzes it, cuts out the subject, writes it a personality, and launch...
Community binary patches for Claude Code CLI — unlock hard-coded limitations via Agent Skills