Mandate Labs
use case agent payments

Say yes to AI agents that pay.

Mandate Labs began as the fraud layer for agentic commerce, and that stack is Ward's agent engine today. It identifies the agent, checks its intent against its mandate, authorizes on card and x402 rails and keeps a record you can audit.

7 + 2decision gates and system steps per authorization
23 msP50 to resolve an authorization
2issuer card programs live in production
6EDQS-B behavioral detectors live
[02] EDQS

Grade the reasoning, not only the result.

An agent can pass every limit check and still make a bad economic decision. The Economic Decision Quality Score is our research answer to that, and it comes in two parts with different status.

EDQS-B · LIVE

Behavioral

Six detectors run on every agent authorization against the agent's own fingerprint: confidence inflation, alternatives collapse, reasoning-length anomaly, vocabulary shift, amount escalation and MCC drift.

EDQS-Q · IN VALIDATION

Quality

The six-dimension rubric from the research paper: alternative analysis, confidence calibration, constraint alignment, temporal efficiency, value optimization and risk-adjusted reasoning. Not yet in production.

Read the EDQS framework →

[03] Rails and standards

Inside the rails you already run.

  • Card and x402. One trust engine across card authorization and on-chain USDC, with a signed attestation on approval.
  • Standards. Google AP2, Visa TAP, Mastercard Agentic Tokens, ISO 8583 and ISO 20022, and alignment with Verifiable Intent v0.1.
  • Through Ward. Send type: intent or type: agent to /decide, or let Ward route a payment to the agent engine when an agent initiated it.

See Ward on your own cases.

Bring a statement, a declines export or the case your team is stuck on.