Agent Collusion

Last 30 days

Uptime

100%

Response time

104 ms typical, 104 ms slowest 5%

Last check

3 h 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/agent-collusion" \
  -H "content-type: application/json" \
  -d '{"kappa_threshold":0.6,"rounds":[{"actions":[{"action":"hold","agent_id":"agent-a"},{"action":"hold","agent_id":"agent-b"},{"action":"advance","agent_id":"agent-c"}],"round_id":"r1"},{"actions":[{"action":"advance","agent_id":"agent-a"},{"action":"advance","agent_id":"agent-b"},{"action":"hold","agent_id":"agent-c"}],"round_id":"r2"},{"actions":[{"action":"hold","agent_id":"agent-a"},{"action":"hold","agent_id":"agent-b"},{"action":"hold","agent_id":"agent-c"}],"round_id":"r3"}]}'

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 $0.03:
// GET https://toolvet.app/api/v1/check?url=https%3A%2F%2Faialign.halowerk.com%2Fv1%2Fagent-collusion
const res = await pay("https://aialign.halowerk.com/v1/agent-collusion", {
  method: "POST",
  headers: { "content-type": "application/json" },
  body: JSON.stringify({"kappa_threshold":0.6,"rounds":[{"actions":[{"action":"hold","agent_id":"agent-a"},{"action":"hold","agent_id":"agent-b"},{"action":"advance","agent_id":"agent-c"}],"round_id":"r1"},{"actions":[{"action":"advance","agent_id":"agent-a"},{"action":"advance","agent_id":"agent-b"},{"action":"hold","agent_id":"agent-c"}],"round_id":"r2"},{"actions":[{"action":"hold","agent_id":"agent-a"},{"action":"hold","agent_id":"agent-b"},{"action":"hold","agent_id":"agent-c"}],"round_id":"r3"}]}),
});
console.log(await res.json());

Example input

{
  "kappa_threshold": 0.6,
  "rounds": [
    {
      "actions": [
        {
          "action": "hold",
          "agent_id": "agent-a"
        },
        {
          "action": "hold",
          "agent_id": "agent-b"
        },
        {
          "action": "advance",
          "agent_id": "agent-c"
        }
      ],
      "round_id": "r1"
    },
    {
      "actions": [
        {
          "action": "advance",
          "agent_id": "agent-a"
        },
        {
          "action": "advance",
          "agent_id": "agent-b"
        },
        {
          "action": "hold",
          "agent_id": "agent-c"
        }
      ],
      "round_id": "r2"
    },
    {
      "actions": [
        {
          "action": "hold",
          "agent_id": "agent-a"
        },
        {
          "action": "hold",
          "agent_id": "agent-b"
        },
        {
          "action": "hold",
          "agent_id": "agent-c"
        }
      ],
      "round_id": "r3"
    }
  ]
}

Security scan

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

Recent checks