agentic-awesome-skills
3,000+ reusable AI agent skills for Claude Code, Codex, Cursor and Gemini CLI. AAS Core adds local catalog discovery, agent-owned selection and plan preview. CL
What is agentic-awesome-skills?
3,000+ reusable AI agent skills for Claude Code, Codex, Cursor and Gemini CLI. AAS Core adds local catalog discovery, agent-owned selection and plan preview. CLI, MCP and plugins.
Overview of agentic-awesome-skills
What it is
Agentic Awesome Skills (AAS) is a library of 2,677+ installable SKILL.md playbooks. AAS Core is the local, agent-first control plane for complete catalog discovery, agent-owned selection, stack validation, and planning. It includes a CLI, local MCP, catalog, plugins, and Workbench. The current release is V19.4.0. It is an independent community project, not affiliated with Google.
Key features
- Local catalog inspection of 2,677+ skills - Agent-owned selection: agents choose exact skills from the catalog - Stack validation: validates chosen IDs and structure in memory - Plan preview: writes an immutable preview for review before any changes - Supports multiple tools: Claude Code, Cursor, Gemini CLI, Codex CLI, and others - Includes CLI, local MCP, catalog, plugins, and Workbench
Use cases
- Discovering and selecting reusable instructions for a project - Inspecting complete skill files before applying them - Keeping an exact skill set that can be reviewed and reused - Validating a stack of skills and previewing a plan before installation
Getting started
Start with AAS Core in Codex or Claude. Configure the local MCP using the provided command: bash npm exec --yes --ignore-scripts [email protected] -- aas mcp configure --host codex --scope user --config /absolute/path/to/codex/config.toml --cache-root /absolute/path/to/aas-cache For direct installation of known skill IDs: bash npm exec --yes --ignore-scripts [email protected] -- agentic-awesome-skills --release 19.4.0 --path .agents/skills --skills brainstorming,systematic-debugging --dry-run Review the preview, then repeat without --dry-run when ready.
License and pricing
License: MIT. Pricing: open source (free).
Based on the project's public README and repository metadata.