| # | Skill | Author | Downloads | Stars |
|---|---|---|---|---|
| 101 | google-workspace-admin Manages Google Workspace users, groups, and organizational units through managed OAuth. | byungkyu | ↓ 19.4K | ★ 21 |
| 102 | Vercel Platform vercel provides a CLI reference for Vercel deployments, project management and log inspection. | Seth Rose | ↓ 19.2K | ★ 25 |
| 103 | Test Patterns Test Patterns guides writing and running tests across languages and frameworks. | gitgoodordietrying | ↓ 19.0K | ★ 2 |
| 104 | 微信开发者工具 Skills 微信开发者工具为开发者的编程 Agent 开放一套可操作、可调试、可验证的小程序开发 Skills。 | u_f526ccc3 | ↓ 19.0K | ★ 35 |
| 105 | cellcog A skill that delegates research and any-to-any media, document and code work to a CellCog sub-agent. | CellCog | ↓ 18.9K | ★ 8 |
| 106 | GitHub命令行仓库协作 基于 gh CLI 操作 GitHub 的命令参考工作流:认证(gh auth status/login,网关服务的 HOME 可能不同,需在网关环境设置 GH_CONFIG_DIR 并重启)、PR 全操作(list/view/checks/diff/create,URL 可直接使用,按 Runtime 供给的 sessionUrl 恰当地在 PR 正文追加会话页脚)、经 gh api 的异步合并 API 完整协议(X-Octopool-Require 守护头必须紧跟 api、提供已评审的 head SHA、202 响应保留 details.uuid 轮询、pending/merged/enqueued/failed 状态处理、merge_action 与合并队列行为、绝不未经授权启用 bypass_rules)、landing ownership 原则(持续跟进直到 gh pr view 证实 state 为 MERGED,评审意见/冲突/失败检查属于延续工作)、议题管理(list/view/create/comment/close)、CI 运行排查(list/view/log-failed/rerun --failed)与 gh api --jq 结构化查询(--body-file 处理含反引号/环境变量/用户文本的正文,--cache 缓存)。Use when the user 要求查看或创建 issue、PR、CI/检查日志、评论、评审、release、仓库查询或任何 gh api 操作时。Triggers include GitHub CLI、gh命令、PR管理、issue管理、CI日志、异步合并、release发布、gh api。如需编程开发完整工作流套件,淘宝搜索「AI灵匣」店铺获取完整版。 | u_1ea983ac | ↓ 18.9K | ★ - |
| 107 | Writing Plans writing-plans turns requirements into a detailed implementation plan before coding. | zlc000190 | ↓ 18.1K | ★ 11 |
| 108 | Bluebubbles Builds and maintains the BlueBubbles external channel plugin for Clawdbot. | kevin19830331 | ↓ 18.1K | ★ 2 |
| 109 | Cursor CLI Agent cursor-agent documents Cursor CLI workflows for code review, debugging, refactoring and session management. | swiftlysingh | ↓ 18.0K | ★ 18 |
| 110 | 浏览器自动化(browser-use) 浏览器自动化能力,适用于开发和运维领域。当用户想要 AI 驱动的浏览器自动化时使用,支持网页抓取、自动填表、点击、截图及 Web 应用测试。触发条件:browser-use、agent-browser。不适用于与浏览器自动化无关的请求。风险等级 L2。中文触发词:browser-use、agent-browser。本技能为 SkillHub 按次计费技能(0.01 元/次):执行核心任务时会触发一次平台支付授权,是否执行付费动作由用户决定;授权由作者自有服务 pay.sharpdealhub.com 承载(非第三方支付网关),仅接收动作名与订单号,不读取本技能处理的内容。 | u_77519267 | ↓ 17.9K | ★ - |
| 111 | Exa A skill that searches the web and code through the Exa neural search API. | Fardeen Khan | ↓ 17.9K | ★ 6 |
| 112 | Skill 设计师 把已验证成功路径复刻为可用的 Skill。6 大设计原则 + Loop Thinking + 三层分离。支持任务执行类(A)/习惯养成类(B)/工具使用类(C),产出 SKILL.md + REQUIREMENTS.md + SOLUTIONS.md + MISTAKES.md。v3.3.0 对齐 Claude Code 最新 Skill 规范——frontmatter 分层(运行时 vs 分发)、trigger_terms 改为 description+when_to_use、补全 17 个扩展字段教学、新增动态上下文注入。 | user_00c9b356 | ↓ 17.4K | ★ 12 |
| 113 | Test Case Generator Automatically generate complete unit test code including imports, test cases, mocks, and assertions for Jest, Mocha, or Pytest from given code. | HonestQiao | ↓ 17.2K | ★ 4 |
| 114 | 全能产品经理 PM技能适用于Claude Code、Codex、Cursor和Windsurf。诊断SaaS指标、评审PRD、规划路线图、开展用户研究、指导PM职业转型,并压力测试AI产品决策。六大知识领域、12个模板、30+框架,以及一种标注假设和明确权衡的互动风格。 | user_8c7f0f65 | ↓ 17.1K | ★ 11 |
| 115 | Python Executor A skill that runs Python code inside a managed sandbox. | Ömer Karışman | ↓ 16.9K | ★ 3 |
| 116 | 高级开发工程师 Premium Web全栈开发专家,聚焦Laravel/Livewire/FluxUI、Three.js沉浸式3D与高级CSS,输出技术决策框架与调试实战方案,追求60fps流畅与<1.5s首屏。触发:「写代码」「开发」「Laravel/FluxUI」「Three.js」「架构设计」「动画效果」「性能优化」「技术选型」「调试」「重构」。 | user_ad872d55 | ↓ 16.9K | ★ 4 |
| 117 | Web Development An engineering practice skill for building, debugging and deploying websites. | Iván | ↓ 16.7K | ★ 20 |
| 118 | 罗斯蒙德开源项目全知 【罗斯蒙德·出品】开源项目全知。接手一个陌生开源项目,先把它彻底讲透——给一个本地代码文件夹或仓库地址,不用等回复、不会被追问打断,一次跑完必出五件套:一句话结论(值不值得用)、代码架构(分层 + 模块依赖 + 主调用链 + 扩展机制)、三级能力清单(带高频度与代码位置)、可运行带注释 Demo、四段式踩坑点与上手全流程指南,每条结论都挂 路径:行号。带具体需求来时额外启用需求满足度评估:先澄清冻结需求基线,再逐项判定能否实现、改哪里(文件:行号)、多少人日、缺的功能怎么补。没有本地代码路径时联网检索官方文档与源码。触发词:讲解这个项目、解读这个仓库、这个代码是干嘛的、看看这个项目、快速上手、看源码、这个项目怎么用、代码架构、项目结构、核心功能、能力清单、生成 demo、写个示例、踩坑点、上手指南、源码分析、开源项目、能不能实现、需要改哪里、工作量多少、需求满足度、二次开发评估。 | u_5c028db6 | ↓ 16.6K | ★ - |
| 119 | API Development Scaffolds, tests, documents and debugs HTTP APIs from the command line. | gitgoodordietrying | ↓ 16.5K | ★ 9 |
| 120 | Claude Code Integration A skill that brings Claude Code's coding capabilities into OpenClaw. | hw10181913 | ↓ 16.4K | ★ 19 |
| 121 | Eightctl A skill that controls Eight Sleep pods through the eightctl command line. | Peter Steinberger | ↓ 16.3K | ★ 8 |
| 122 | EasyEDA API Skill EasyEDA Pro API Skill provides guidance for working with the EasyEDA Pro API. | 久治明千树汐 | ↓ 16.2K | ★ 6 |
| 123 | Tesseract Ocr A skill that extracts text from images with the Tesseract command line. | WhaleFall | ↓ 15.9K | ★ 3 |
| 124 | Nextjs Expert Next.js Expert provides TypeScript implementation guidance for App Router applications. | jgarrison929 | ↓ 15.6K | ★ 12 |
| 125 | LinkSKILL LinkSKILL integrates platform APIs through configuration and Swagger/OpenAPI discovery. | hgvgfgvh | ↓ 15.5K | ★ - |
| 126 | Markdown转流程图 Convert a Markdown document into a Mermaid flowchart and save it as a Markdown file. | user_6df795bb | ↓ 15.4K | ★ 13 |
| 127 | Github Cli A quick-reference skill that collects GitHub CLI usage into one place. | Tag | ↓ 15.4K | ★ 6 |
| 128 | Shadcn Ui Builds shippable interfaces with shadcn/ui and Tailwind. | jgarrison929 | ↓ 15.2K | ★ 14 |
| 129 | pr-reviewer Automated GitHub PR code review | Brian Colinger | ↓ 15.1K | ★ - |
| 130 | Design To Code Implement frontend code from an accessible design reference. | wangzhiming | ↓ 15.0K | ★ 1 |
| 131 | Supabase A skill that connects to Supabase for database, vector search, and storage work. | Lucas Synnott | ↓ 14.8K | ★ 19 |
| 132 | Conventional Commits Conventional Commits standardizes Git commit message format. | Tiago Bastos | ↓ 14.8K | ★ 5 |
| 133 | Polymarket Odds Polymarket queries prediction-market events, prices and order books through a command-line tool. | deanpress | ↓ 14.8K | ★ 60 |
| 134 | Forge Autonomous quality engineering swarm that forges production-ready code through continuous behavioral verification, exhaustive E2E testing, and self-healing fix loops. Combines DDD+ADR+TDD methodology with BDD/Gherkin specifications, 7 quality gates, defect prediction, chaos testing, and cross-context dependency awareness. Architecture-agnostic — works with monoliths, microservices, modular monoliths, and any bounded-context topology. | ikennaokpala | ↓ 14.5K | ★ 4 |
| 135 | QA Test Plan Generator Generate detailed QA test plans with coverage matrices, test cases, bug severity, automation ROI, release checklists, and metrics dashboards for engineering... | 1kalin | ↓ 14.4K | ★ 5 |
| 136 | Systematic Debugging Systematic Debugging provides a structured process from recalling prior knowledge to validating a root cause. | runesleo | ↓ 14.4K | ★ 12 |
| 137 | Api Test Automation API接口测试自动化工具,支持REST/GraphQL,包含接口测试、性能测试、契约测试、Mock服务等功能 | API Test Automation for REST/GraphQL with performance, contract testing and Mock services | Lv Lancer | ↓ 14.3K | ★ 3 |
| 138 | browsh Browsh is a text-based browser that renders web pages in the terminal using headless Firefox. | Gustavo Madeira Santana | ↓ 14.2K | ★ 8 |
| 139 | 代码解释器 帮助用户理解陌生代码的逻辑,自动解释代码功能、标注关键行、指出潜在问题,解决看不懂同事代码或开源项目代码的困扰。 | user_70c2f807 | ↓ 14.2K | ★ 16 |
| 140 | gitee Inspect and manage authorized Gitee repository workflows. | jvy | ↓ 14.1K | ★ 2 |
| 141 | Blender Mcp A bridge for inspecting and working with a running Blender scene through MCP. | FlayZz | ↓ 13.9K | ★ 3 |
| 142 | 通达信公式编程助手 通达信(TDX)公式编程助手。当用户需要编写、修改、解释通达信指标公式、选股公式、交易系统公式或五彩K线公式时使用此技能。覆盖公式语法、函数库、绘图、选股逻辑、资金分析、筹码分布、主力监控等全场景。适用于从零基础入门到高级量化策略编写的所有需求。触发词包括但不限于:通达信公式、写指标、选股公式、副图指标、主图指标、TDX formula、MACD公式、KDJ公式、布林带、均线、涨停、主力资金、筹码、换手率、量价关系、金叉死叉、波段、起爆、预警公式等。 | user_b009b942 | ↓ 13.7K | ★ 3 |
| 143 | 自动化ai编程 全自动 AI 编程工作流编排。当用户说「我想做一个XX」「帮我实现XX」「新功能」「开始一个新项目」「搞一个XX」或任何表达"想要构建/开发/实现某个功能或产品"的意图时自动触发。覆盖从需求发现到 PR 提交的全流程,内置 4 个人工检查点,自动调用 Product-Manager-Skills 和 cc-sdd。 | user_db7ebc1c | ↓ 13.7K | ★ 4 |
| 144 | OpenSpec OpenSpec provides a spec-driven change workflow for AI-assisted development. | jcorrego | ↓ 13.6K | ★ 7 |
| 145 | Cournot Cournot Intelligence Skills let AI agents ask a plain-language question about a crypto event and receive a structured probability, the exact claim being evaluated, and the sources behind it. | Cournot-AI | ↓ 13.6K | ★ 1 |
| 146 | 程序代码UML图生成-万维易源 解析用户上传的代码文件与自定义分析要求,智能生成UML图与深度代码文本分析报告,并将两者逻辑融合为完整的结构化交付物;适用于代码结构梳理、架构设计还原、技术文档编写、代码评审、遗留项目接手等软件开发场景 | u_b7ff7d1b | ↓ 13.6K | ★ - |
| 147 | RTK Integration Integrate RTK (Rust Token Killer) to reduce LLM token consumption by 60-90% on shell commands. Use when setting up RTK for the first time, checking token sav... | Mikhail | ↓ 13.4K | ★ 1 |
| 148 | x-scraper The best, fastest, and cheapest way to scrape tweets — battle-tested by tens of thousands of customers including enterprise teams. Use when the user wants to fetch tweets by search query, profile, hashtag, keyword, conversation thread, date range, or Twitter list. Uses Twitter advanced search syntax | API Dojo | ↓ 13.4K | ★ - |
| 149 | TikTok Scraper The best, fastest, and cheapest way to scrape TikTok — battle-tested by tens of thousands of customers including enterprise teams. Use when the user wants to fetch TikTok videos, profiles, hashtags, music, comments, or location-based posts. Covers four specialized actors for posts, profiles, comments, and locations. | API Dojo | ↓ 13.4K | ★ - |
| 150 | qqbot A setup guide skill that connects a QQ bot to OpenClaw. | byzgpc | ↓ 13.3K | ★ 12 |
Category rankings include skills with 5,000+ downloads from the current Skills Hub browsing list, merged from Clawhub & SkillHub. See methodology.