Huqan

Huqan is a deterministic governance and verification layer designed for AI-assisted work. It evaluates agent claims and actions against evidence, provenance, context, and policy while protecting canonical memory from unsupported writes and routing risky operations through explicit approval workflows.

The system records consequential decisions with inspectable Trust Receipts for auditability. Huqan operates as a trust boundary layer rather than an LLM or universal truth engine, enabling organizations to implement verifiable oversight of AI agent behavior.