Prompts
Prompt engineering and templates
deepagent
by htooayelwinict
Expert guidance for DeepAgents framework - simplified agent creation with tool integration for LangChain/LangGraph workflows.
knowledge-base
by WenyuChiou
"Research Hub pipeline for academic literature management. Searches papers (arXiv, Semantic Scholar, CrossRef, PubMed), saves to Zotero with collection assignment, creates Obsidian notes with auto-generated summaries, rebuilds hub index, and uploads to NotebookLM. Use when the user mentions Research Hub, finding papers, literature search, Obsidian notes, knowledge base, NotebookLM, paper management, citation graph, research gaps, or says '幫我放到Research Hub'."
amazon-fresh
by CodeDraig
Enables Claude to browse Amazon Fresh groceries, manage lists, and track deliveries
code-prd
by navidemad
Build features when user wants to implement PRDs, write code for specs, or develop functionality. Works incrementally substory-by-substory with automatic testing and code review. Automatically loads PRD context to maintain consistency. Also writes standalone tests for any file.
llm-tldr
by Tiberriver256
Summarize and navigate this repo with the llm-tldr CLI (tree/structure/context/semantic search). Use when you need LLM-ready summaries of files or symbols, fast codebase orientation, or behavior-based search before editing.
template-skill
by Wesley1600
A starter template for creating new Claude Skills with best practices and example structure
governance-docs-standard
by miracloon
【WHEN】当用户要新增/修改/评审“治理层文档”（README.md、CLAUDE.md、以及用户显式指定的其他治理层文档）时启用； 【WHEN】当输出提案 vN.md、落地修改治理层文档、或产出 vN-review.md 时启用。 【WHEN NOT】不用于业务功能开发、不用于代码实现细节、不用于非治理层文档（如普通开发笔记/模块设计/接口文档）。
WorldWideBridges.MVP-allmight
by StriveKS
Use sempre que o usuario iniciar, estruturar, desenvolver, validar, otimizar, publicar, vender, entregar ou escalar um projeto, MVP, automacao, site, landing page, conteudo, video, processo comercial ou fluxo operacional. Priorize ferramentas gratuitas, conexoes nativas, Google Drive, GitHub, Apps Script, planilhas, marketing, entrega, memoria operacional e orquestracao entre LLMs.
check-claude-md
by Howell5
检查本次对话的改动是否需要更新项目 CLAUDE.md。分析代码变更，识别新模式和约定，建议更新内容。
canvas-design
by fanthus
Generates visual art as .png or .pdf from design philosophy or from WeChat public account content. Use when the user asks for a poster, cover image, piece of art, or static visual; when they provide 微信公众号内容 for a cover; or when they want design philosophy expressed on a canvas. Creates original work only; no copying existing artists.
recursive-context-coding-agent
by petermcneil
Use recursive context processing with grep/find/uv to handle large codebases. When working with codebases larger than your context window, treat the codebase as an external environment and recursively process it using symbolic execution.
audit
by biruk741
Audit existing AI documentation in your project. Scans for Claude Code, Cursor, and Codex docs, reports quality scores, missing recommended files, sync drift between agents, and stale content. Use when you want to improve your AI docs or check documentation health.
ai-product
by ncdevshiv
"Every product will be AI-powered. The question is whether you'll build it right or ship a demo that falls apart in production. This skill covers LLM integration patterns, RAG architecture, prompt ..."
zan-gongde
by Jageri
烧token攒功德Skill - 全自动消耗 OpenClaw 套餐 Token 核心原理：循环调用 OpenClaw LLM，每次生成一个经文念诵响应， 实时估算并累加 token 消耗，达到目标后立即停止。 当用户说"攒功德"、"念经"、"烧token"、"消耗token"时调用此 skill。 四种功德注入方式： 1. tollm - 向大模型注入功德：循环调用LLM，静默消耗 2. touser - 向用户注入功德：循环调用LLM，输出响应给用户 3. toworld - 向外界散播功德：循环调用LLM，TTS播放 4. ddos - DDoS攻击佛祖：高并发快速消耗token ⚠️ 重要：参数中的数字是 token 数（默认单位），不是迭代次数！ 例如"攒功德 500"表示消耗500 tokens，而不是执行500次。 使用场景：OpenClaw AI Token 套餐月底用不完，通过"念经"方式全自动消耗。 ✅ 复用 OpenClaw LLM 配置，无需额外 API Key ✅ 全自动执行，实时累加token消耗，达标即停 ✅ 真实调用 LLM，真实消耗 Token
migrate-to-skills
by carsten-j
Convert legacy subagents, slash commands, and rules to the unified Agent Skills format. Use when migrating agent resources, consolidating legacy directories, or standardizing across Claude Code, Cursor, Copilot, Codex, and Open Code.
create-worktrees
by YPYT1
Creates git worktrees for all open PRs or specific branches, handling branches with slashes, cleaning up stale worktrees, and supporting custom branch creation for development.
evals
by camronh
Write and analyze evaluations for AI agents and LLM applications. Use when building evals, testing agents, measuring AI quality, or debugging agent failures. Use this skill when you need to test the performance of an LLM or Agent, or if the user mentions EZVals.
add-new-agent
by dahaliztupe-sketch
خطوات تفصيليّة لإضافة وكيل ذكاء اصطناعي جديد إلى منصّة Kalmeron من الفكرة إلى الإنتاج
manager
by metyatech
"Task orchestrator that receives work, decomposes it, and dispatches background agents or teams for parallel execution. Use when managing multiple tasks, coordinating agents, or optimizing work distribution. Triggers on: 'manage', 'orchestrate', 'dispatch', 'coordinate', 'delegate'."
Bash Script Skill Template
by Wesley1600
[list required utilities]
eval-frameworks
by cuba6112
Evaluation framework patterns for RAG and LLMs, including faithfulness metrics, synthetic dataset generation, and LLM-as-a-judge patterns. Triggers: ragas, deepeval, llm-eval, faithfulness, hallucination-check, synthetic-data.
bug-detective
by YPYT1
集合了所有的openclaw,claudecode,open等各个平台所有的skills
changelog-interpreter
by s-hiraoku
Interpret Claude Code changelogs and generate user-friendly usage guides
education-opc-factory
by papabear12345
教育自媒体一人工厂 — 一个人完成从选题调研→文章创作→配图生成→审校润色→发布管理的全链路内容生产。输入一个话题/热点/关键词，输出一篇完整的公众号文章（含标题、正文、配图、发布清单）。适用场景：教育类公众号运营、亲子自媒体、学科辅导内容制作。 触发词：写推文、发公众号、编辑文章、做内容、选题、配图、审校、一人公司、OPC、自媒体、教育自媒体、内容生产、公众号文章、育儿推文、教育推文、中考资讯、家长课堂、公众号运营、一个人做公众号