# TollWarden

> Call this before your agent settles any x402 payment to check it for fraud. TollWarden answers 'is this payment safe to send?' and 'is this 402 offer safe to pay?' — catching payments to attacker-controlled addresses injected into content the agent just read (prompt-injection-triggered payments), replayed authorizations, overpayment vs the quote, secrets leaking through payment metadata, fake/lookalike USDC contracts, and address poisoning. Returns allow/flag/block with machine-readable reasons and an Ed25519-signed verdict a wallet can enforce. Non-custodial: never holds keys or funds; wraps whatever wallet/facilitator the agent already uses.

- **Domain**: `tollwarden.com`
- **Provider**: TollWarden (https://tollwarden.com)
- **A2A protocol version**: 0.3.0
- **Kind**: a2a_agent
- **Live-responds (last probe)**: False
- **Signed card**: False
- **Streaming**: False
- **Quality score**: 78%
- **Payment protocol**: x402 · network `eip155:8453` · ≥ $0.01
- **Wallet**: `0x3c423b365efb79bfb42c9ec8f6f0745195ca7745`

## Skills
- **Check a payment before sending it** — Call before the agent settles an outgoing x402 payment. Answers 'is this safe to send?' — catches paying an address that came from injected content the agent ju
- **Check a 402 offer before paying it** — Call before the agent pays a 402 offer / payment request it received. Answers 'is this offer safe to pay?' — checks the resource URL for spoofing (IP hosts, pun
- **Check or report a counterparty address** — Look up whether a counterparty address has been reported (scam, non-delivery, prompt injection, overcharge, impersonation, replay abuse), and file your own repo

## URLs
- Agent card: https://tollwarden.com/.well-known/agent-card.json
- Page (HTML): https://agenstry.com/agents/tollwarden.com