by Ezeafk
Curated reusable skills, workflows, and tool-backed capabilities for AI agents.
# Add to your Claude Code skills
git clone https://github.com/Ezeafk/awesome-agent-skillsGuides for using ai agents skills like awesome-agent-skills.
awesome-agent-skills is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by Ezeafk. Curated reusable skills, workflows, and tool-backed capabilities for AI agents. It has 81 GitHub stars.
awesome-agent-skills's catalog security scan is still queued. You can run an instant dependency and prompt-injection check now with the "Scan for vulnerabilities" button above.
Clone the repository with "git clone https://github.com/Ezeafk/awesome-agent-skills" and add it to your Claude Code skills directory (see the Installation section above).
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh awesome-agent-skills against similar tools.
No comments yet. Be the first to share your thoughts!
Unlocks once the catalog security scan passes (runs nightly).
The deep catalog scan for this skill is still queued. Run an instant dependency check now instead.
A curated list of reusable skills, workflows, and tool-backed capabilities for AI agents.
Languages: English | 简体中文 | Português do Brasil | 日本語 | Deutsch | Русский
This list focuses on reusable capabilities that help AI agents complete real tasks. It is not a prompt dump. A listed project should have a clear use case, reusable structure, installation or usage instructions, and enough context to explain inputs, outputs, and safety boundaries.
Last reviewed: 2026-06-28
For this list, a skill is a reusable capability package for an AI agent. It may include instructions, workflows, scripts, references, templates, MCP servers, connectors, local app automation, validation steps, or examples.
| Field | Values |
|---|---|
| Platform | Codex, Claude, ChatGPT, Cursor, Copilot, MCP, Generic Agent, n8n |
| Includes | Workflow, scripts, references, templates, MCP server, connector, app integration, examples |
| Status | Active = recent maintenance; Stable = still useful with slower updates; Use with care = valuable but sensitive |
| Risk | Low = mostly read-only or content generation; Medium = files, APIs, accounts, browser automation; High = production, security, deletion, payments, credentials, bulk actions |
Skills are scored before inclusion. Stars are a weak signal only; a niche skill can be included when it solves a real task, has reusable structure, and gives agents enough instructions or tools to run safely.
| Area | Points | What reviewers check |
|---|---|---|
| Task clarity | 0-2 | Clear real-world use case, inputs, outputs, and target user. |
| Reusable structure | 0-2 | More than a prompt: skill files, workflows, templates, scripts, examples, or references. |
| Platform and tool fit | 0-2 | Clear support for Codex, Claude, ChatGPT, Cursor, Copilot, MCP, n8n, CLI, browser, API, or local app workflows. |
| Validation and examples | 0-2 | Demo, tests, sample usage, CI checks, schema, or step-by-step verification. |
| Maintenance and safety | 0-2 | Recent activity, license clarity, safe defaults, confirmation rules, and credential/privacy guidance. |
Inclusion guide: 8-10 = recommended, 6-7 = acceptable with caveats, 4-5 = use with care or re-review, below 4 = remove or do not include. High-risk workflows need a stronger score and explicit safety notes.
| Skill | Platform | Use case | Includes | Status | Risk |
|---|---|---|---|---|---|
| anthropics/skills | Claude, Generic Agent | Public repository for Agent Skills. | Skills, examples, references | Active | Low |
| github/awesome-copilot | Copilot, Generic Agent | Community instructions, agents, skills, and configurations. | Instructions, agents, skills, configs | Active | Low |
| tech-leads-club/agent-skills | Claude Code, Cursor, Copilot, Codex | Secure validated skill registry for professional coding agents. | Skill registry, validation, metadata | Active | Medium |
| taishi-i/awesome-ChatGPT-repositories | ChatGPT, Codex, Claude | Searchable curated AI repository list exposed through Claude Code skills. | Awesome list, search workflow, skills | Active | Low |
| JSONbored/awesome-claude | Claude, MCP, Generic Agent | Registry of Claude workflow assets, agents, MCP servers, skills, commands, and hooks. | Registry, agents, skills, MCP, commands | Active | Medium |
| msitarzewski/agency-agents | Claude Code, Copilot, Cursor, Codex, Gemini CLI, OpenCode, Aider, Windsurf | Multi-role AI agency library with installable specialists for coding, security, content, product, and business workflows. | Agent library, install scripts, converters, examples | Active | Medium |
| Skill | Platform | Use case | Includes | Status | Risk |
|---|---|---|---|---|---|
| openai/codex | Codex | Terminal coding agent for repository tasks. | CLI, sandboxed workflows, code automation | Active | Medium |
| anthropics/claude-code | Claude Code | Agentic coding in the terminal with repo context and git workflows. | CLI, tool workflows, coding automation | Active | Medium |
| anthropics/claude-code-action | Claude Code, GitHub | Run Claude Code workflows from GitHub Actions. | GitHub Action, automation workflow | Active | Medium |
| jabrena/cursor-rules-java | Cursor, Claude, Codex, Copilot | AI-native Java enterprise development workflow. | Rules, agents, commands, MCP servers | Active | Medium |
| DollarDill/beads-superpowers | Claude Code, Codex, OpenCode | Process-discipline skills and persistent task memory for coding agents. | Skills, task memory, TDD workflows | Active | Medium |
| vosslab/vosslab-skills | Claude, Codex | Refactoring, code review, repo maintenance, and education content workflows. | Reusable skills, workflows | Active | Medium |
| avalonreset/legends-github | Claude Code, Codex, Gemini CLI | GitHub repository optimization and community health workflows. | Skills, audit workflow, metadata checks | Active | Medium |
| Jakevin/fastcontext-agent-tools | Codex, MCP | Repository exploration with FastContext. | MCP server, Codex skill | Active | Medium |
| Skill | Platform | Use case | Includes | Status | Risk |
|---|---|---|---|---|---|
| Kaelio/ktx | Claude, Codex, MCP | Executable context layer for analytics agents. | Semantic context, database querying, MCP | Active | Medium |
| dagster-io/dagster | Generic Agent | Data asset orchestration for agent-driven analytics workflows. | Pipelines, schedules, observability | Active | Medium |
| EtienneLescot/n8n-as-code | n8n, Claude, Cursor, Copilot | Give agents n8n workflow creation and sync capabilities. | TypeScript workflows, schemas, templates | Active | Medium |
| product-on-purpose/pm-skills | Claude, Codex | Product management skill families and structured product workflows. | Skills, templates, examples, CI contracts | Active | Medium |
| LLMQuant/awesome-trading-agents | Generic Agent, MCP | Market research, strategy, and trading agent skills. | Awesome list, MCP servers, workflows | Active | High |
| Skill | Platform | Use case | Includes | Status | Risk |
|---|---|---|---|---|---|
| yusufkaraaslan/Skill_Seekers | Claude | Convert docs, GitHub repos, and PDFs into Claude skills. | Parser, OCR, conflict detection, MCP | Active | Medium |
| teng-lin/notebooklm-py | Claude, Codex, Python | Programmatic NotebookLM workflows for agents. | Python API, CLI, agentic skill | Active | Medium |
| Skill | Platform | Use case | Includes | Status | Risk |
|---|---|---|---|---|---|
| LingyiChen-AI/comfyui-workflow-skill | Claude Code, Cursor, Generic Agent | Generate ComfyUI workflow JSON from natural language. | Templates, node definitions, model workflows | Active | Medium |
| SlavaSexton/ComfyUI-Agent-Kit | Claude Code, Codex, Gemini CLI | Drive local ComfyUI end to end from an agent. | Skill, prompt recipes, templates, automation | Active | Medium |
| Skill | Platform | Use case | Includes | Status | Risk |
|---|---|---|---|---|---|
| microsoft/playwright-mcp | MCP | Browser automation through Playwright MCP. | MCP server, browser tools | Active | Medium |
| executeautomation/mcp-playwright | MCP, Claude, Cursor | Browser and API automation through Playwright. | MCP server, Playwright workflows | Active | Medium |
| browserbase/mcp-server-browserbase | MCP | Let agents control cloud browsers with Browserbase and Stagehand. | MCP server, browser control | Active | Medium |
| soxoj/awesome-osint-mcp-servers | MCP | OSINT tool discovery for research agents. | Curated MCP server list | Active | High |
| Skill | Platform | Use case | Includes | Status | Risk |
|---|---|---|---|---|---|
| NVIDIA/SkillSpector | Claude, Codex, Generic Agent | Scan AI agent skills for malicious patterns, unsafe instructions, and security risks |