theodosia

msradam/theodosia
★ 0 stars Python 🤖 AI/LLM Updated 1d ago
Put an AI agent on rails: mount a Burr state machine as an MCP server so the agent can only take the next allowed step, with every step recorded and replayable.
View on GitHub → Try with Claude — $10 free →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "theodosia": {
      "command": "uvx",
      "args": [
        "theodosia"
      ]
    }
  }
}

Or install with pip: pip install theodosia

README Excerpt

**Theodosia puts an AI agent on rails.** You define a workflow once as a [Burr](https://burr.dagworks.io/) state machine, and Theodosia serves it over [MCP](https://modelcontextprotocol.io/) so the agent can only take the next allowed step, with every step recorded and replayable. The model can be wrong; the model cannot lie about state.

Tools (2)

coffeeupstream

Topics

agent-orchestrationapache-burrfastmcpllm-agentsmcpmodel-context-protocolobservabilitypythonstate-machinetheodosia