Clearance402 · Trust layer for x402 & MCP agent services
x402 lets agents pay. Clearance402 tells them what is safe to pay for. Verify paid x402/MCP tools in real time — live probes, price checks, output verification, and risk scoring — before delegated wallet permissions are spent.
Discovery and payments already exist. The missing wedge is trust: developers need to prove their tools are legit, agents need a machine-readable safe/unsafe decision, and users need a human-readable clearance card before approving spend.
Run live probes, confirm a real 402 flow, evaluate output with Venice, and generate a Trust Card with proof and integration snippets.
ExploreCall Clearance402 before paying a tool and receive ALLOW, WARN, BLOCK, RETEST, or HUMAN_APPROVAL_REQUIRED — with reasons.
ExploreHuman-readable clearance cards, MetaMask Advanced Permission spend limits, and one-click revoke before delegated spending happens.
ExplorePassed live probe, price check, permission scope, output quality, and risk checks.
Works, but has latency, low confidence, new listing, price drift, or limited history.
Failed the x402 flow, returned mismatched output, exceeded scope, or looks risky.
Status is stale or recent behavior changed — run another probe before payment.
Risk or spend is high enough that a person must approve manually.
Probe endpoints, confirm a real 402 challenge, pay, retry, and verify the returned output against the declared schema.
Eight dimensions: protocol, price, output, reliability, permission safety, relayer readiness, drift, and dev readiness.
Venice scores output quality, behavior drift, and risk labels as part of the main clearance flow.
Scout discovers, Buyer pays, Verifier evaluates, Guardian blocks or allows — coordinated around each call.
MetaMask Advanced Permissions (ERC-7715) spend caps and revocation, checked before redemption.
Every probe, payment, block, approval, relay, and evaluation is written to the audit log.
© 2026 Clearance402. All rights reserved.