agentic-awesome-skills

MCP.Pizza Chef: sickn33

Lets your AI coding assistant search a library of more than 2,000 ready-made skill guides and pick out exactly the ones that fit your project. Instead of you hunting through documentation on your own, your assistant looks at your project, chooses matching skills, and puts together a reviewable plan before anything is changed. Works with Claude, Cursor, Codex CLI, and Goose, among other AI coding tools. No account or API key is required to browse and select skills.

Coding

Use This MCP server To

Find the right coding skill guides for my project automatically Search a library of over 2,000 skill playbooks for a specific task Build a reviewable plan of chosen skills before anything changes Compare several skill options before picking one for a task See which skills are already set up in my project Review my assistant's skill choices in a browser tool before applying them Spot gaps where no matching skill exists yet for my project

README

AAS Core — Agentic Awesome Skills

Local, agent-owned skill stacks for coding agents—from complete catalog access to a reproducible, reviewable plan.

Current release: V15.8.0. This release includes AAS Core for complete local catalog search, agent-owned selection, manifest validation, planning, and diagnosis. Apply and recovery remain experimental and outside the supported preview path.

Codex or Claude inspects your project and chooses exact skills from the complete local AAS catalog. AAS Core does not rank or recommend them: its read-only compose_stack tool validates the agent-owned selection in memory, and a client or the aas CLI can persist it as aas-stack.json and produce an immutable plan before any target change.

Read the AAS Core preview guide →

Project
  -> inspected by Codex or Claude (not by AAS)
  -> agent searches and reads the complete local catalog
  -> AAS MCP (local stdio, read-only)
  -> Codex or Claude chooses exact skill IDs
  -> compose_stack validates the selection in memory (read-only)
  -> client or AAS CLI persists aas-stack.json and optional evidence
  -> AAS CLI validate + immutable plan preview
  -> human review (optionally in Workbench)

The reusable SKILL.md playbooks, specialized plugins, bundles, workflows, and direct installers remain important. They are the content, curation, distribution, and compatibility layers around AAS Core—not competing primary products.

This is an independent community project. It is not affiliated with, sponsored by, endorsed by, or authorized by Google. Google, Antigravity, Gemini, and related product names are referenced only to describe compatibility and install targets. The GitHub repository is canonical; the hosted catalog and browser-local Workbench are companion discovery and review surfaces, not a hosted control plane.

agentic-awesome-skills FAQ

Can I use this to find the best skill for a specific coding task?
Yes — your AI assistant searches the full catalog of more than 2,000 skill guides, compares the options, and picks the ones that best match what your project needs.
Can I use this to review my project's setup for missing skills?
Yes — the assistant is guided to look across your whole project, from features to testing to security, and flag areas where no matching skill exists yet.
Which apps does this work in?
It is built to work with Claude, Cursor, Codex CLI, and Goose, and it also supports several other AI coding tools.
Do I need an API key or account to use it?
No account or API key is required to browse the catalog, search skills, or build a plan.
How hard is it to set up?
Setup is light — you add a small connection snippet to your AI assistant's settings, and the assistant takes care of the rest.
Will it change my project files automatically?
No — by default it only searches, selects, and plans. Actually applying any changes is a separate, optional step you have to turn on yourself.
Can I review what my assistant picked before anything happens?
Yes — every selection is saved as a plan file you can look over, and there is also a browser-based review tool called Workbench.
Does this work with assistants built on OpenAI, Claude, or Gemini?
Yes — it is designed to work across different AI assistants, including tools built on Claude, OpenAI's Codex, and Google's Gemini CLI.