# ActiveInferenceInstitute/pymcp

Source: https://glimind.com/server/github%2Factiveinferenceinstitute%2Fpymcp
Registry id: `github/activeinferenceinstitute/pymcp`
Status: unknown

## Tools (22)

- `calculate_free_energy`: Calculate free energy for an agent.
- `create_agent`: Create an active inference agent from a model dict (A, B, optional C, D).
- `create_grid_world_env`: Create a grid world environment with rewards.
- `create_gridworld_agent`: Gridworld agent with position + reward observation models.
- `define_generative_model`: Build random A/B matrices + default C/D from dimension lists.
- `get_agent`: Retrieve stored agent as JSON.
- `get_all_agents`: List all stored agents.
- `get_all_environments`: List all stored environments.
- `get_all_functions`: List all public methods on the interface.
- `get_all_simulations`: List all stored simulations.
- `get_environment`: Retrieve stored environment as JSON.
- `infer_policies`: Optimize beliefs over policies via expected free energy.
- `infer_states`: Infer hidden-state posterior for an observation (FPI, BP, …).
- `infer_states_from_observation`: Infer states from an observation.
- `reset_environment`: Reset an environment to its initial state.
- `run_simulation`: Run a full agent–environment simulation loop.
- `sample_action`: Sample an action from the policy posterior.
- `step_environment`: Advance an environment given an action.
- `validate_generative_model`: Validate a generative model.
- `visualize_agent_model`: Visualize agent model.
- `visualize_belief_dynamics`: Visualize belief dynamics.
- `visualize_simulation`: Visualize simulation results.

Measured by Glimind (https://glimind.com): MCP handshake + tool list only, no tool is ever called. CC BY 4.0.