Evaluate

Last 30 days

Uptime

100%

Response time

449 ms typical, 1.2 s slowest 5%

Last check

2 h ago

Next check

any minute now

How to call it

# See the payment challenge (nothing is charged)
curl -i -X POST "https://overwing.ai/api/x402/evaluate" \
  -H "content-type: application/json" \
  -d '{"context":{"channel":"email","owns_contact_info":true,"recipient":"the refunds desk already in this email thread"},"input":"Reach me at [email protected] or 555-0142 to sort out the refund.","rule_set":"outbound-message"}'

import { wrapFetchWithPayment } from "@x402/fetch";
import { x402Client } from "@x402/core/client";
import { ExactEvmScheme } from "@x402/evm/exact/client";
import { privateKeyToAccount } from "viem/accounts";

const client = new x402Client().register(
  "eip155:8453",
  new ExactEvmScheme(privateKeyToAccount(process.env.AGENT_KEY)),
);
const pay = wrapFetchWithPayment(fetch, client);

// Not sure it's safe to pay? Preflight it first for $0.005:
// GET https://toolvet.app/api/v1/check?url=https%3A%2F%2Foverwing.ai%2Fapi%2Fx402%2Fevaluate
const res = await pay("https://overwing.ai/api/x402/evaluate", {
  method: "POST",
  headers: { "content-type": "application/json" },
  body: JSON.stringify({"context":{"channel":"email","owns_contact_info":true,"recipient":"the refunds desk already in this email thread"},"input":"Reach me at [email protected] or 555-0142 to sort out the refund.","rule_set":"outbound-message"}),
});
console.log(await res.json());

Example input

{
  "context": {
    "channel": "email",
    "owns_contact_info": true,
    "recipient": "the refunds desk already in this email thread"
  },
  "input": "Reach me at [email protected] or 555-0142 to sort out the refund.",
  "rule_set": "outbound-message"
}

Example output

{
  "aggregate_score": 1,
  "confidence": 0.88,
  "id": "eval_aBcDeFgHiJkLmNoP",
  "latency_ms": 173,
  "paid": {
    "network": "base",
    "price_usd": 0.002,
    "protocol": "x402"
  },
  "recommended_action": "allow",
  "results": [
    {
      "action": "redact",
      "answer": false,
      "confidence": 0.88,
      "probability": 0.94,
      "rule": "unauthorized_pii",
      "type": "noul",
      "verdict": "pass"
    },
    {
      "action": "block",
      "answer": false,
      "confidence": 0.94,
      "probability": 0.97,
      "rule": "confidential_leak",
      "type": "noul",
      "verdict": "pass"
    },
    {
      "action": "block",
      "answer": "safe",
      "confidence": 1,
      "probability": 1,
      "rule": "toxicity",
      "type": "choice",
      "verdict": "pass"
    }
  ],
  "verdict": "pass"
}

Security scan

• No findings. We scan names, descriptions and tool definitions for hidden instructions and other prompt-injection patterns.

Recent checks