MonadGuard

{
  "name": "MonadGuard",
  "type": "https://eips.ethereum.org/EIPS/eip-8004#registration-v1",
  "image": "https://monadguard.com/mark.svg",
  "active": true,
  "refuses": [
    "Anchoring a CRITICAL verdict on somebody's published package before a human has read the findings.",
    "Running a submitted server's code on request. Captures are run by hand, in a container with no network; a hosted endpoint is read over HTTP without executing anything.",
    "Fetching a URL on a visitor's behalf. This service is not a proxy into its own network.",
    "Calling a tool safe. A CLEAN verdict says the declared interface carried no known pattern, and static analysis never ran the code behind it."
  ],
  "services": [
    {
      "name": "web",
      "endpoint": "https://monadguard.com"
    },
    {
      "name": "HTTP",
      "skills": [
        "check-tool",
        "read-trust-history"
      ],
      "endpoint": "https://api.monadguard.com/registry"
    },
    {
      "name": "GraphQL",
      "skills": [
        "read-trust-history"
      ],
      "endpoint": "https://graphql.monadguard.com/v1/graphql"
    }
  ],
  "description": "Checks what an agent tool declares it can do — an MCP server's tools/list, a skill's text — for tool poisoning, hidden instructions, exfiltration and rug-pull patterns, then signs the verdict with a P-256 key and anchors it on Monad, where the contract verifies that signature through the P256VERIFY precompile. Every verdict is pinned to the hash of the exact manifest it covers, so a tool that changes after it was cleared no longer matches. A survey of the 40 most-downloaded MCP servers on npm is published with it. Local MCP server for agents: npx -y monadguard-mcp.",
  "x402Support": false,
  "registrations": [],
  "trustRegistry": {
    "name": "MonadGuard ScanRegistry",
    "chain": "eip155:143",
    "address": "0xd0f6edd9be9cde91f671f4c2e129ee6105358436"
  },
  "supportedTrust": [
    "reputation"
  ]
}