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.
**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.