
Write HTML. Render video. Built for agents.
Cheshire Cat AI is a bottom-up framework for understanding, extending, and running AI agents as a service. It exposes a local UI and REST API, and the README says it is built to be easy to extend via vibe coding and agentic engineering. It also supports plugins and agent-style integrations.
Builders who want a Python agent framework they can run, extend, and study.
You can spin up an agent service locally and extend it with your own behavior instead of starting from scratch.
Starts a local agent app with a UI and REST API playground.
Provides a framework designed to be extended with new agent behavior and integrations.
Supports plugin-style additions for changing how the agent works.
Is positioned as a learning and research base for understanding agent systems.
uv init --bare uv add cheshire-cat-ai uv run ccat
[!WARNING] Version 2 is unstable alpha! This new version is still under heavy development and needs work before it becomes stable. Expect breaking changes, rough edges, and incomplete features. Not recommended for production use yet.
Best framework to learn how AI agents work.
It is designed bottom up to be easy to understand and easy to extend via vibe coding / agentic engineering.
Main use case of the Cat is education & research. Many companies and institutions use it to publish agents on the web.
Give the project a star ⭐ if you like creative cats!
In an empty folder:
uv init --bare
uv add cheshire-cat-ai
uv run ccat
Roadmaps are for amateurs.
If you have a proposal, open an issue and only do a PR if the issue is assigned to you. Read the contributor guide.
Code is licensed under GPL3.
The Cheshire Cat AI logo and name are property of Piero Savastano (founder and maintainer).
"Would you tell me, please, which way I ought to go from here?"
"That depends a good deal on where you want to get to," said the Cat.
"I don't much care where--" said Alice.
"Then it doesn't matter which way you go," said the Cat.
(Alice's Adventures in Wonderland - Lewis Carroll)
Sign in to join the discussion.
No comments yet. Be the first to say what this is good for.

Write HTML. Render video. Built for agents.
Ultra-lightweight, open-source, self-hosted personal AI agent framework in Python with WebUI, tools, memory, MCP, multi-agent workflows, automation, and chat apps
SkillOpt is a text-space optimizer that trains reusable natural-language skills for frozen LLM agents through trajectory-driven edits, validation-gated updates, and deployable best_skill.md artifacts.

Omnigent is an open-source AI agent framework and meta-harness: orchestrate Claude Code, Codex, Cursor, Pi, and custom agents — swap harnesses without rewriting, enforce policies and sandboxing, and collaborate in real time from any device.
A theoretical reconstruction of the Claude Mythos architecture, built from first principles using the available research literature.
🕷️ An adaptive Web Scraping framework that handles everything from a single request to a full-scale crawl!