MolTrust is now A2A v0.3 conformant
Any A2A-compatible agent or orchestrator can now discover MolTrust's trust verification capabilities through standard A2A agent discovery — no custom integration required.
What changed
We updated our Agent Card at /.well-known/agent-card.json to fully conform to the A2A v0.3 specification. The card declares five agent skills:
- trust-score — W3C DID-based trust score (0–100) with behavioral history and on-chain proof
- did-resolution — W3C DID Document resolution with verification methods and service endpoints
- credential-verification — AAE delegation chain validation and VC verification
- wallet-binding — Cryptographic DID-to-wallet binding verification (EVM + Solana)
- sybil-detection — Endorsement-graph clustering and behavioral anomaly detection
The card also exposes a MolTrust-specific extension (moltrust.ch/extensions/trust-score/v1) that tells A2A clients exactly how to query trust scores and integrate them into their agent workflows.
Why it matters
A2A's Agent Card is how agents discover each other's capabilities. Until now, MolTrust was reachable via API — but invisible to the A2A ecosystem by default. With a conformant Agent Card, any A2A orchestrator can find us, understand what we do, and integrate trust verification into agent-to-agent interactions without reading our documentation first.
The A2A spec has authorization schemes on its roadmap. When that lands, MolTrust will be the trust layer that A2A agents reach for to answer: “Should I trust this agent, based on its history?”
Try it
Trust score for any registered DID:
TechSpec v0.8.1: moltrust.ch
API: api.moltrust.ch