{"x402Version":2,"name":"PaySafe","description":"Check a payment for fraud before your agent sends it. Call PaySafe before settling any x402 payment to avoid the ways agent wallets get drained: paying an attacker-controlled address injected into web or tool content the agent just read (prompt-injection-triggered payments), replayed payment authorizations, paying more than the quoted price, leaking secrets or private keys through payment metadata, fake/lookalike USDC token contracts, and address poisoning. Returns allow/flag/block with machine-readable reasons; verdicts are Ed25519-signed so a wallet can enforce them. Advisory and non-custodial — never touches keys or funds. Includes a shared counterparty reputation registry (free to report to).","resources":[{"resource":"https://paysafe-agent.com/v1/scan/outgoing","type":"http","x402Version":2,"accepts":[{"scheme":"exact","network":"eip155:8453","amount":"10000","asset":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","payTo":"0x3C423B365efB79bFB42C9eC8f6F0745195ca7745","maxTimeoutSeconds":60}],"metadata":{"description":"Is this payment safe to send? Screen an outgoing x402 payment BEFORE settling it. Catches payments to an address that came from content the agent just read (prompt-injection-triggered payments), replayed nonces, overpayment vs the quoted price, secrets/PII leaking in payment metadata, fake/lookalike USDC contracts, address poisoning, velocity anomalies, and known-bad or reported counterparties. Returns allow/flag/block with per-check reasons and an Ed25519-signed verdict."}},{"resource":"https://paysafe-agent.com/v1/scan/incoming","type":"http","x402Version":2,"accepts":[{"scheme":"exact","network":"eip155:8453","amount":"10000","asset":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","payTo":"0x3C423B365efB79bFB42C9eC8f6F0745195ca7745","maxTimeoutSeconds":60}],"metadata":{"description":"Is this 402 offer safe to pay? Screen an incoming x402 payment request BEFORE paying it. Checks the resource URL for spoofing (IP-literal hosts, punycode/homoglyphs, link shorteners, credential demands like 'send your seed phrase'), price sanity, replay, and whether the counterparty has been reported. Returns allow/flag/block with reasons."}},{"resource":"https://paysafe-agent.com/v1/reputation/{address}","type":"http","x402Version":2,"accepts":[{"scheme":"exact","network":"eip155:8453","amount":"10000","asset":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","payTo":"0x3C423B365efB79bFB42C9eC8f6F0745195ca7745","maxTimeoutSeconds":60}],"metadata":{"description":"Has anyone reported this address? Check a counterparty before dealing with it: aggregated post-hoc reports (scam, non-delivery, prompt injection, overcharge, impersonation, replay abuse) filed by other agents, with distinct-reporter counts and a risk level."}}],"free_tier":"First 100 calls free per API key — POST /v1/keys (no payment) to get one.","reporting":"POST /v1/reputation/report is always free, so agents can flag bad counterparties after the fact.","custody":"non-custodial; advisory verdicts only","plans":{"catalog":"https://paysafe-agent.com/v1/plans","summary":"Subscription tiers with lower per-scan pricing and higher velocity/spend headroom. Agents can subscribe and renew autonomously: POST /v1/plans/subscribe is itself x402-paid at the plan price. Safety-critical checks are identical on every tier."}}