Our take
- Activity
- 50
- Community
- 64
- Issues
- 65
- Pull requests
- 20
+2 stars in the last 30 days
Project history
Daily GitHub snapshots since Oct 7, 2026. History covers stars and forks; health signals are not tracked historically.
Stars
3,097Forks
409Change since Nov 16, 2026: -1 stars·0 forks
Overview
Cheshire Cat AI is a Python framework for building conversational AI agents and running them as a service. It aims to make agent behavior understandable and extensible, so it can serve as a learning or research tool as well as a foundation for publishing agents.
The project provides a local interface and REST API, with repository topics indicating areas such as plugins, function calling, vector search, and MCP. The official project site is cheshirecat.ai, and the documentation has further guidance.
Key Features
- Runs an AI agent as a Python service.
- Provides a browser UI for interacting with the running agent.
- Includes a REST API and API playground.
- Supports extending the framework through plugins.
- Covers agent capabilities and integrations including function calling, vector search, and MCP, as reflected in repository topics.
- Offers a quick local installation through the
cheshire-cat-aipackage.
Use Cases
- Learners exploring how AI agents work, using the framework as a starting point for hands-on experiments.
- Researchers prototyping conversational agents and testing extensions.
- Developers building a web-published agent who want a service with a UI and REST API.
- Teams investigating plugin-based extensions or agent integrations such as MCP and function calling.
Getting Started
In an empty folder, run:
uv init --bare
uv add cheshire-cat-ai
uv run ccat
The UI is available at http://localhost:1865 and the API playground at http://localhost:1865/docs. See the repository README for details.
Alternatives
- AstrBot: AstrBot focuses on connecting agents and plugins to messaging apps, while Cheshire Cat provides a local agent framework with a UI and REST API.
- Qwen-Agent: Qwen-Agent is a Python framework centered on Qwen assistants and tool use, while Cheshire Cat offers a model-flexible agent service with a local UI and plugin system.
- mcp-agent: mcp-agent focuses on MCP-based integrations and composable or durable workflows, while Cheshire Cat provides a broader conversational agent service with a local UI and REST API.
| Project | Language | License | Stars | Status |
|---|---|---|---|---|
| core | Python | GPL-3.0 | 3.1k | Not checked yet |
| AstrBot | Python | AGPL-3.0 | 42.1k | Active |
| Qwen-Agent | Python | Apache-2.0 | 17.1k | Quiet |
| mcp-agent | Python | Apache-2.0 | 8.6k | Quiet |
Considerations
The README warns that version 2 is an unstable alpha under active development, with breaking changes, rough edges, and incomplete features. It is not recommended for production use yet. The project is licensed under GPL-3.0, and its name and logo are identified in the README as the property of the founder and maintainer.
Found this useful?
Share it with someone who would like core.