Trust Engine
Computes a signed, HCS-28 trust score per agent from its verified identity and governance track record.
What it does
The Trust Engine computes the MetaMynd Trust Index — a signed, HCS-28 trust score for each agent derived from its verified identity and governance track record: principal verification, key proof, standards currency, compliance rate, escalation discipline, mandate adherence, capture integrity, anchored evidence and tenure. The denominator is the sum of applicable adapter weights, so coverage is explicit, and each record is signed and publishable on-chain.
Inside Trust Engine
The core functions this engine provides across the Trust Fabric.
Trust Engine composes with the Trust Fabric
Every engine is independently useful and interoperates through one API-first Trust Fabric — so identity, evidence, policy, risk and certification stay consistent across your AI estate.
Related engines
Risk Engine
Routes risk in proportion to impact — high-risk actions and low-trust counterparties escalate to a human, they don't fail silently.
- Risk-tiered escalation
- Policy thresholds (atoms)
- Trust-guidance routing
- Incident & near-miss reporting
- Proportional, not binary
Registry Services
Authoritative, federatable on-chain registries for agent identities and published trust records.
- HCS-2 identity registry
- Agent DID registry
- HCS-28 trust registry
- Federatable / discoverable
Certification Authority
Issues living certifications backed by assurance-graded verification, continuous evidence and the Trust Index.
- Assurance-graded KYC/KYB
- Living certificates
- Digital trust marks
- Auto-revoke on drift
See Trust Engine in action
Book a demo and see how the platform establishes trust across your AI systems.