active-inference

What problem does it solve?

Active Inference helps an agent decide what to do next when the true state is hidden and there is no reliable scalar reward signal to guide exploration.

Core Features & Use Cases

• Bayesian decision-making without rewards: selects actions by minimizing Expected Free Energy, balancing epistemic value (information gain) and pragmatic value (preference satisfaction).
• Generative model for partially observable environments: uses likelihood (A), transition (B), preferences (C), and priors (D) with variational inference to update beliefs from observations.
• Principled tool selection and debugging workflows: drives probing actions (e.g., searches, targeted reads, test runs) based on which action most reduces uncertainty before committing to fixes.

Quick Start

Use this skill to build a small active-inference agent that models hidden causes of a bug and chooses the next tool action (like searching a file or running tests) that most reduces diagnostic uncertainty until you reach a confident conclusion.