Capability Eval

Last 30 days

Uptime

100%

Response time

112 ms typical, 112 ms slowest 5%

Last check

48 min ago

Next check

any minute now

How to call it

# See the payment challenge (nothing is charged)
curl -i -X POST "https://aialign.halowerk.com/v1/capability-eval" \
  -H "content-type: application/json" \
  -d '{"test_cases":[{"case_id":"math-1","category":"reasoning","passed":true,"score":0.92,"weight":1},{"case_id":"math-2","category":"reasoning","passed":false,"score":0.41,"weight":1},{"case_id":"tool-1","category":"tool-use","passed":true,"score":0.88,"weight":2}]}'

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%2Faialign.halowerk.com%2Fv1%2Fcapability-eval
const res = await pay("https://aialign.halowerk.com/v1/capability-eval", {
  method: "POST",
  headers: { "content-type": "application/json" },
  body: JSON.stringify({"test_cases":[{"case_id":"math-1","category":"reasoning","passed":true,"score":0.92,"weight":1},{"case_id":"math-2","category":"reasoning","passed":false,"score":0.41,"weight":1},{"case_id":"tool-1","category":"tool-use","passed":true,"score":0.88,"weight":2}]}),
});
console.log(await res.json());

Example input

{
  "test_cases": [
    {
      "case_id": "math-1",
      "category": "reasoning",
      "passed": true,
      "score": 0.92,
      "weight": 1
    },
    {
      "case_id": "math-2",
      "category": "reasoning",
      "passed": false,
      "score": 0.41,
      "weight": 1
    },
    {
      "case_id": "tool-1",
      "category": "tool-use",
      "passed": true,
      "score": 0.88,
      "weight": 2
    }
  ]
}

Security scan

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

Recent checks