正在加载项目…
正在加载项目…
内容来自 GitHub 原始文件，由原作者维护。在 GitHub 查看
Unlike many bulk-generated skill repositories, this collection focuses on real-world Agent Skills created and used by actual engineering teams, not mass AI‑generated stuff.
Compatible with Claude Code, Codex, Antigravity, Gemini CLI, Cursor, GitHub Copilot, OpenCode, Windsurf, and more. See the table below for paths and documentation.
The most contributed Agent Skills repository, built and maintained together with the community.
 | TestMu AI (formerly LambdaTest) is an AI-native testing cloud platform built for modern engineering teams. Covering everything from autonomous test creation and fast execution to testing AI agents, chatbots and voice assistants. |
 | Crawlbase is web data infrastructure trusted by 70,000+ developers. Its Crawling API, MCP server, and integrations give AI agents live access to any webpage — with JavaScript rendering, proxy rotation, and anti-bot protection. |
 | SerpApi is a Web Search API for your AI apps. Available in Markdown and JSON for any integration. |
You ship products with AI, but every launch still dies quietly because nobody posts about it. EveryFeed plugs your AI assistant into a social workspace that drafts, schedules, and publishes across 35+ channels — no agency, no marketing hire.
Official skills by VoltAgent for building AI agents with the VoltAgent TypeScript framework.
Official skills by the SerpApi team — the Web Search API for AI apps. They give agents structured, machine-readable search data through 130+ engines, from Google web and Scholar to Maps, Flights, Hotels, and Shopping.
More from SerpApi (not skills, but they pair with them):
Official skills by the Crawlbase team that give AI agents live web access through the Crawlbase MCP server: crawl any URL as raw HTML, clean Markdown, or a screenshot, and manage results in Crawlbase Cloud Storage.
More from Crawlbase (not skills, but they pair with them):
Production-grade Agent Skills for every major test automation framework, maintained by the TestMu AI (formerly LambdaTest) team. They help AI coding assistants generate expert-level test automation code across web, mobile, API, BDD, and unit testing stacks.
/scrape for single-page extraction from product code/search for query-first discovery with optional content hydrationOfficial skills by Venice.ai for the Venice API.
Agent Skills for the Stitch MCP server, compatible with Claude Code, Gemini CLI, Cursor, and more.
Official Google Workspace CLI skills for managing Google Workspace services via the gws CLI tool.
Official AI agent skills from the Expo team for building, deploying, and debugging Expo apps.
Official AI agent skills from the Hugging Face team for ML workflows.
Domain-specific knowledge for Azure SDK and Microsoft AI Foundry development. 133 skills across 6 languages.
Official curated skills from OpenAI's skills repository.
Official skills from Figma's MCP server guide.
Official marketing skills by Corey Haines, covering the full SaaS marketing stack from SEO and copywriting to growth, CRO, and paid acquisition.
Direct-response advertising skills by Kim Barrett, organized into foundations, copy-chief, operator-os, orchestrators, and QA — covering avatar work, offer design, Schwartz-style copywriting, creative testing, and full-funnel campaign orchestration.
Official Web3 and trading skills from the Binance team. Includes crypto market data, on-chain analytics, token security auditing, and spot trading via the Binance API.
Official skills from the Apollo GraphQL team for building GraphQL clients, servers, federated supergraphs, and the Apollo Router.
Official authentication and identity skills from the Auth0 team. Covers SDKs for popular frameworks plus workflows for MFA, migration, and quickstart detection.
Official skills from the Brave team for accessing Brave Search APIs including web, images, videos, news, and local point-of-interest data.
Official browser automation skills from the Browserbase team. Covers headless browsing, cookie sync, serverless functions, and adversarial UI testing.
Official AI code review skills from the CodeRabbit team.
Official wallet, payments, and trading skills from the Coinbase team. Covers USDC transfers, onchain queries, x402 paid APIs, and Base trading.
Observability skills from Datadog Labs for APM, logs, monitors, and LLM observability workflows powered by the pup CLI.
Official skills from the Firebase team for setup, auth, Firestore, hosting, Genkit AI SDKs, and security rule auditing.
Official skills from the Flutter team covering layouts, state, navigation, native interop, platform setup, and testing for cross-platform Flutter apps.
46 battle-tested product management skills by Dean Peters. Frame problems, hunt opportunities, scaffold validation experiments, and kill bad bets fast — with frameworks from Teresa Torres, Geoffrey Moore, Amazon, MITRE, and more.
Component Skills
65 product management skills by Paweł Huryn, creator of The Product Compass newsletter. Covers the full PM lifecycle — from discovery and strategy to execution, analytics, and go-to-market — with frameworks from Teresa Torres, Geoffrey Moore, and more.
Data Analytics
Execution
11 development and document generation skills from the MiniMax AI team. Covers frontend, fullstack, mobile, shader development, document creation (PDF, DOCX, XLSX, PPTX), and the MiniMax AI CLI with MiniMax API integration.
Official DuckDB skills for data querying, file reading, and documentation search directly from Claude Code.
Official GSAP animation skills covering the full GreenSock ecosystem — core API, timelines, ScrollTrigger, plugins, utilities, React integration, performance optimization, and framework support.
28 skills by Garry Tan (Y Combinator CEO) that transform Claude Code into a virtual engineering team — structured workflows from ideation to production deployment. Ships 600K+ lines of production code in 60 days.
Official skills from Notion's repositories — workspace-aware skills for capturing knowledge, preparing meetings, researching, and turning specs into tasks.
From notion-cookbook:
From claude-code-notion-plugin:
Official Resend skills to send and receive emails, build email templates and power your agent with email expertise.
Lighthouse-style web quality skills from Addy Osmani(Google Chrome team) covering performance, Core Web Vitals, accessibility, SEO, and modern best practices — the same pillars Google Lighthouse audits.
Official MongoDB Agent Skills for agentic workflows — connection management, schema design, query optimization, natural language querying, and Atlas Stream Processing.
Official skills published by NVIDIA for its AI, accelerated computing, robotics, simulation, and developer platforms. NVIDIA updates and reorganizes this catalog frequently, so this list links to the maintained source instead of duplicating a snapshot.
Official Google Cloud skills covering Firebase, BigQuery, Cloud Run, GKE, AlloyDB, Cloud SQL, Gemini Enterprise Agent Platform, networking observability, and the Well-Architected Framework. 19 skills.
Extend the power of AI across your organization with a curated library of skills, agents, and MCP servers—all backed by your Red Hat subscription. Whether you’re an SRE optimizing workflows in Cursor or an architect building intelligent interfaces, leverage trusted building blocks to deploy and scale agentic automation with confidence.
redhat/cve-skillpack - Understand CVEs, check product lifecycle status, gather diagnostics, and file support cases at the right severity — essential Red Hat skills for everyday operations.
redhat/sre-skillpack - Discover, remediate, and verify CVEs across your RHEL fleet — orchestrating Red Hat Lightspeed and Ansible Automation Platform through a single workflow.
redhat/openshift-skillpack - Provision, inventory, and report on OpenShift clusters — spanning Assisted Installer, OCM, ROSA, ARO, and kubeconfig fleets — through a single conversational workflow.
redhat/openshift-virtualization - Manage the full VM lifecycle on OpenShift Virtualization — create, clone, snapshot, restore, rebalance, and report — through a single conversational workflow.
Official skills published by Cypress to help create, maintain, understand, and fix your Cypress tests. 3 skills.
VoDaiLocz/kilo-kit-mcp - Comprehensive library of 177 curated skills paired with an MCP runtime enforcing protocol-level C4 workflow gates, hard-gated command execution with security guardrails, and 5 cognitive reasoning engines (Tree of Thoughts DAG, Adversarial Grilling, 5-Whys Root Cause Tracer, Context Compactor, Self-Evolution) for Claude Code, Cursor, Antigravity, and Codex
hedralab/eskill - Meta-skill to build top-tier Agent Skills: spec-compliant SKILL.md, eval loop, validator, market research, numbered-file pipeline
robzolkos/skill-rails-upgrade - Analyze Rails apps and provide upgrade assessments
antonbabenko/terraform-skill - Terraform and OpenTofu patterns: testing, modules, state, CI/CD.
zxkane/aws-skills - AWS development with infrastructure automation and cloud architecture patterns
Rootly-AI-Labs/rootly-incident-responder - AI-powered incident response with ML similarity matching, solution suggestions, and on-call coordination. Requires Rootly MCP Server
conorluddy/ios-simulator-skill - Control iOS Simulator
ramzesenok/iOS-Accessibility-Audit-Skill - Audit iOS App against Accessibility norms
truongduy2611/app-store-preflight-skills - Scan iOS/macOS projects to catch common mistakes that lead to App Store rejection before submission
- Code review and PR autofix workflows for coding agents
pypdfium2), real homework quiz banks, hard exit-code anti-hallucination, and persistent cross-session state. Tested on 1,000+ pages of real college courseware on small cheap models.Skills in this list are curated, not audited. They may be updated, modified, or replaced by their original maintainers at any time after being added here.
Before installing or using any Agent Skill, review potential security risks and validate the source yourself.
Recommended tools:
Agent skills can include prompt injections, tool poisoning, hidden malware payloads, or unsafe data handling patterns. Always review the code and use skills at your own discretion.
| Tool | Project Path | Global Path | Official Docs |
|---|---|---|---|
| Antigravity | .agents/skills/ | ~/.gemini/config/skills/ | Antigravity Skills |
| Claude Code | .claude/skills/ | ~/.claude/skills/ | Claude Code Skills |
| Codex | .agents/skills/ | ~/.agents/skills/ | Codex Skills |
| Cursor | .cursor/skills/ | ~/.cursor/skills/ | Cursor Skills |
| Gemini CLI | .gemini/skills/ | ~/.gemini/skills/ | Gemini CLI Skills |
| GitHub Copilot | .github/skills/ | ~/.copilot/skills/ | Copilot Skills |
| OpenCode | .opencode/skills/ | ~/.config/opencode/skills/ | OpenCode Skills |
| Windsurf | .windsurf/skills/ | ~/.codeium/windsurf/skills/ | Windsurf Cascade Skills |
| mblode/agent-skills | Nobody ships AI slop on purpose. These skills make sure you don't. UI audits, typography, docs, PR review, and releases. npx skills add mblode/agent-skills |
As the ecosystem grows, consistent quality helps agents discover and use skills reliably. The following references and criteria keep the bar high.
| Area | Guideline |
|---|---|
| Description | Write in third person. State what the skill does and when to use it. Use specific keywords agents can match on (e.g., "PostgreSQL migration" not "database stuff"). |
| Progressive disclosure | Keep top-level metadata under ~100 tokens. Skill body should stay below 500 lines. Load resources (large docs, schemas) on demand, not inline. |
| No absolute paths | Never hard-code machine-specific paths like /Users/alice/. Use relative paths or well-known variables ($HOME, $PROJECT_ROOT). |
| Scoped tools | Request only the tools the skill actually needs. Avoid blanket "tools": ["*"]. Declare tool dependencies explicitly. |
We welcome contributions! See CONTRIBUTING.md for guidelines.
Note: Please don't submit skills you created 3 hours ago. We're now focusing on community-adopted skills, especially those published by development teams and proven in real-world usage. Quality over quantity.
MIT License - see LICENSE
This is a curated list. Skills listed here are created and maintained by their respective authors and teams, not by us. We select community-adopted, proven skills and do not audit, endorse, or guarantee the security or correctness of listed projects. They are not security-audited and should be reviewed before production use.
If you find an issue with a listed skill or want your skill removed, please open an issue and we'll take care of it promptly.
| Paweł Huryn | MiniMax | DuckDB | GSAP |
| Garry Tan (gstack) | Notion | Resend | Addy Osmani (Web Quality) |
| MongoDB | Kim Barrett (Advertising) | Apollo GraphQL | Auth0 |
| Brave | Browserbase | CodeRabbit | Coinbase |
| Datadog Labs | Firebase | Flutter | Venice.ai |
| Red Hat | Community | Redis | NVIDIA |
| Google Cloud | Quality Standards |
@crawlbase/mcpcharacter_slug usageInteractive Skills
Workflow Skills
Go-to-Market
Market Research
Marketing & Growth
Product Discovery
Product Strategy
Toolkit
sanjay3290/postgres - Execute safe read-only SQL queries against PostgreSQL databases
sanjay3290/deep-research - Autonomous multi-step research using Gemini Deep Research Agent
jthack/ffuf-claude-skill - Web fuzzing with ffuf
lackeyjb/playwright-skill - Browser automation with Playwright
woniu9524/open-web-bridge - Drive your real, logged-in Chrome via CDP from Claude Code, Codex, or Gemini CLI: semantic snapshots, real mouse clicks, human handoff for captchas and logins, HAR capture and replay
ibelick/ui-skills - Opinionated, evolving constraints to guide agents when building interfaces
muthuishere/hand-drawn-diagrams - Generate hand-drawn Excalidraw diagrams from a prompt — animated SVG, hosted edit link, and PNG export. Works with Claude Code, Codex, Gemini CLI, and any agent supporting standard skill paths
nextlevelbuilder/ui-ux-pro-max-skill - UI/UX design patterns and best practices
ehmo/platform-design-skills - 300+ design rules from Apple HIG, Material Design 3, and WCAG 2.2 for cross-platform apps
Kayforkind/reimagine-it - Redesign HTML pages using only their existing content
scarletkc/vexor - Vector-powered CLI for semantic file search with a Claude/Codex skill
obra/test-driven-development - Write tests before implementing code
obra/subagent-driven-development - Development using multiple sub-agents
obra/systematic-debugging - Methodical problem-solving in code
obra/finishing-a-development-branch - Complete Git code branches
obra/requesting-code-review - Initiate code review processes
obra/receiving-code-review - Process and incorporate code feedback
obra/using-git-worktrees - Manage multiple Git working trees
obra/verification-before-completion - Validate work before finalizing
obra/writing-skills - Develop and document capabilities
fvadicamo/dev-agent-skills - Git and GitHub workflow skills for commits, PRs, and code reviews
omkamal/pypict-skill - Pairwise test generation
alinaqi/maggy - Opinionated project initialization with security-first guardrails, spec-driven atomic todos, LLM testing patterns, and CLI tool orchestration (gh, vercel, supabase)
ZhangHanDong/makepad-skills - Makepad UI development skills for Rust apps: setup, patterns, shaders, packaging, and troubleshooting.
massimodeluisa/recursive-decomposition-skill - Handle long-context tasks (100+ files, 50k+ tokens) through recursive decomposition strategies based on RLM research
AvdLee/swiftui-expert-skill - Modern SwiftUI best practices and iOS 26+ Liquid Glass adoption
efremidze/swift-patterns-skill - Modern Swift/SwiftUI best practices
wendylabsinc/claude-skills - Swift Server development guidance with linting tool for best practices
rorkai/app-store-connect-cli-skills - Automate App Store deployments and management using ASC CLI
rameerez/claude-code-startup-skills - Skills for building and running software startups, apps, and SaaS
zscole/model-hierarchy-skill - Cost-optimized model routing based on task complexity
CloudAI-X/threejs-skills - Three.js skills for creating 3D elements and interactive experiences
Leonxlnx/taste-skill - High-agency frontend skill that gives AI good taste with tunable design variance, motion intensity, and visual density to stop generic UI slop
testdino-hq/playwright-skill - 70+ production-tested Playwright automation testing patterns: E2E, POM, CI/CD, migrations, CLI
NeoLabHQ/review - Comprehensive PR code review using specialized agents: bug-hunter, security-auditor, code-quality-reviewer, contracts-reviewer, historical-context-reviewer, test-coverage-reviewer
NeoLabHQ/reflexion - Self-refinement loop that forces the LLM to reflect on previous output and correct itself.
NeoLabHQ/sdd - Spec-driven development workflow that transforms prompts into production-ready implementations through structured planning, architecture design, and LLM-as-a-Judge based quality gates.
NeoLabHQ/ddd - Domain-driven development skills that also include Clean Architecture, SOLID principles, and design patterns.
NeoLabHQ/sadd - Dispatches independent subagents for individual tasks with code review checkpoints between iterations for rapid, controlled development.
NeoLabHQ/kaizen - Applies continuous improvement methodology with multiple analytical approaches, based on Japanese Kaizen philosophy and Lean methodology.
uucz/moyu - Anti-over-engineering skill with 5 variants and 10 platforms
mattpocock/skills - 17 dev workflow skills: PRD writing, TDD, codebase architecture, git guardrails, issue triage, refactoring plans, and more
mukul975/Anthropic-Cybersecurity-Skills - 753 cybersecurity skills across 38 domains: cloud security, pentesting, red teaming, DFIR, malware analysis, threat intel, and more (MITRE ATT&CK mapped)
wrsmith108/varlock-claude-skill - Secure environment variable management ensuring secrets are never exposed in Claude sessions, terminals, logs, or git commits
Skill_Seekers - Automatically convert documentation websites, GitHub repositories, and PDFs into Claude AI skills in minutes
NoizAI/skills - Human-like TTS workflows with local/cloud APIs and app delivery
Kevin7Qi/codex-collab - Collaborate with Codex from Claude Code
ethos-link/rails-conventions - Rails 8 conventions for consistent production code changes
mcollina/skills - 11 skills by Matteo Collina: Node.js, Fastify, TypeScript, OAuth, Git/GitHub, ESLint neostandard, documentation (Diataxis), Node.js core internals, skill optimizer, and more
Lum1104/understand-anything - Interactive codebase knowledge graphs via multi-agent LLM analysis
hqhq1025/skill-optimizer - Diagnose and optimize Agent Skills (SKILL.md) with real session data and research-backed static analysis. Works with Claude Code, Codex, and any Agent Skills-compatible agent
LambdaTest/agent-skills - TestMu AI (Formerly LambdaTest) Skills is a curated collection of Agent Skills that teach AI coding assistants how to write production-grade test automation.
foryourhealth111-pixel/Vibe-Skills - A skills governed plug-and-play harness for staged, test-driven skill orchestration
metalbear-co/skills - Skills that let agents code and test against your Kubernetes cluster using mirrord
dembrandt/dembrandt-skills - UX and design system skills: hierarchy, typography, accessibility, interactions
GanyuanRan/Aegis - Evidence-driven method pack for AI coding agents
baskduf/codex-fable5 - Evidence-based workflow gates for Codex
csthink/dashmotion - Animated technical diagrams from plain English or Mermaid, self-contained HTML/SVG
plasma-ai/fractal - Bounded hierarchical agent loops in isolated git worktrees
reliefeai/browser-relay - Control an existing logged-in Chrome without stealing focus
squirrelscan/squirrelscan - Audits websites for SEO, performance, security, accessibility and returns fixes
Simon-He95/markstream-install - Install streaming Markdown renderers across five frontend frameworks
eduardo-sl/go-agent-skills - Curated Go skills for code review, concurrency, testing, and architecture
drogers0/github-image-upload - Attach screenshots, PDFs, logs, zips, and videos to GitHub PRs, issues, and comments, returning canonical user-attachments URLs. GitHub has no public attachment-upload API. Works with Claude Code, Codex, Cursor, and Gemini CLI
browser-act/browser-act - Automate authenticated browsers with extraction and human handoff
agiwhitelist/auteur - Builds websites gated by an anti-slop linter and motion QA
JasonColapietro/suede-creator-skills - Design, UI polish, code review grading, AI evals, SEO audits.
superdesigndev/superdesign-skill - Creates design systems from existing codebases and iterates UI drafts
Ryan-yang125/motion-lexicon - Build and review product motion with installable React components
Maksim-Burtsev/simple-man - Strips praise, recaps and filler from agent answers while keeping every fact you act on: findings carry location and fix, refusals carry the safe procedure, tutorials stay long-form. Benchmarked on 1,793 preregistered live calls with raw records committed. Works with Claude Code, Codex, Gemini CLI, Cursor
aeonfun/aeon - 70+ Claude Code skills + autonomous GitHub Actions agent framework
KhazP/vibe-coding-prompt-template - Plan MVPs into PRD, tech design, and AGENTS.md
lindblomstefan/skills-library - Guided discovery skill for Claude Code: runs an interview to recommend from a catalog of 100+ AI skills; records session feedback that validates candidates over time
rainmanjam/poka-yoke - Make misuse unrepresentable: audit, design, and enforce mistake-proofing devices
scarletkc/agents - Reusable standards and workflow skills for AI coding agents
dannwaneri/spec-writer - Turns vague requests into spec, plan, and tasks
Continuum-AI-Corp/orca-replay - Answer questions about past agent runs from their recordings
tt-a1i/archify - Generate validated interactive architecture diagrams from codebases or system descriptions
d1vai/d1v - Deploy web projects with verified previews and confirmed production releases
kensaurus/cursor-kenji - Ready-made playbooks your coding agent auto-triggers
saleh-alhaddad/itqan-engineering - Full software-engineering lifecycle in 12 skills: a resumable orchestrator plus spec, plan, TDD build, verify, five-axis review, security, and release — with approval gates before code and evidence before "done"
lukstei/slop-grader - Rule-based CLI tool powered by TypeSafe Jev that evaluates documents against custom rulesets, producing document scores and line-by-line violation flags to guide auto-fixing with an AI agent
mfs-find / mfs-ingest skills that search, grep and read across your code, docs, chat (Slack/Gmail/Jira), databases and object stores as one file-like, searchable namespace; self-hosted with local ONNX embeddings