DOCUMENTATION

The trust layer for agent payments.

Fidacy verifies the agent, checks the mandate against the user's signed intent, scores the risk, and issues a cryptographically signed verdict — then keeps a tamper-proof record of it. It fills the risk_data container in AP2 and rides natively on A2A, so AI agents can move money and you can prove who authorized what.

THE PROBLEM

The protocols that carry agent payments — AP2, A2A — leave a deliberate gap: a place to attach a risk decision, but no opinion on who fills it. Without a trust layer, a merchant or PSP receiving an agent-initiated payment has no way to know whether the agent is who it claims to be, whether it stayed inside the user's mandate, or who to hold accountable in a dispute. Fidacy is that trust layer.

What happens on every request

A single call to POST /v1/assess runs the full pipeline and returns in milliseconds:

{{ p.n }}
{{ p.title }} {{ p.body }}

The verdict

Every assessment returns a flat decision plus a rich, versioned outcome object — risk level, the consolidated reason, per-dimension confidences, typed signals, and pointers to the signed payload and the audit entry.

POST /v1/assess · 200 
{
  "decision": "approve",            // approve | review | deny
  "score": 12,
  "assessmentId": "asmt_…",
  "riskPayloadJws": "eyJhbGciOiJFZERTQSJ9…",
  "outcome": {
    "outcome": { "decision": "approve", "risk_level": "normal" },
    "confidences": { "agent_identity": 1, "mandate_validity": 1 },
    "risk_payload": { "alg": "EdDSA", "verify_url": "…/.well-known/jwks.json" },
    "audit": { "entry_id": "ae_…", "sequence": 41 }
  }
}
THE SAFETY INVARIANT

Any error, timeout, or fault degrades the decision to review — never approve. Anonymous or hard-violating traffic can never come back approved. This holds across the entire system, including the asynchronous layers.

Standards & compliance

AP2

Fills the risk_data container in Payment Mandates.

A2A

Version negotiation, Agent Card verification, verdict as Task.metadata.

EU AI ACT

The append-only, hash-chained audit trail is the non-repudiable evidence Article 10-class systems require.

EdDSA / JWS

Every risk payload is a signed JWS, verifiable by any party against the public JWKS.

Where to go next

The manual — install, first session, precautions, troubleshooting→ Quickstart — your first verified payment in three calls→ How Fidacy works — principals, agents, mandates, policies, trust, audit→ A2A protocol — negotiate, verify the Agent Card, return Task.metadata→ API reference — every endpoint, field, and error→