MCP Tools

Model Context Protocol (MCP) tool definitions for integrating TrustUCP with AI agents. These tools enable agents to perform trust lookups, verify proofs, and evaluate policies.

About MCP

What is Model Context Protocol?

MCP is an open protocol that allows AI agents to interact with external tools and data sources. TrustUCP provides MCP-compatible tools that let AI agents query trust status, verify cryptographic proofs, and enforce policies when processing transactions.

Learn about MCP →

Available Tools

evaluate_policy

MCP Tool

Evaluate a tenant policy using business verification signals and checkout context.

Input Schema

{
  "type": "object",
  "additionalProperties": false,
  "required": [
    "tenant_id"
  ],
  "properties": {
    "tenant_id": {
      "type": "string"
    },
    "business_verification": {
      "$ref": "../../../../jsonschema/trust/business-verification/v1/schema.json"
    },
    "context": {
      "type": "object",
      "additionalProperties": true
    }
  }
}

Output Schema

{
  "$ref": "../../../../jsonschema/trust/trust-decision/v1/schema.json"
}

get_evidence_pack

MCP Tool

Fetch a JSON evidence pack by tenant and pack identifier.

Input Schema

{
  "type": "object",
  "additionalProperties": false,
  "required": [
    "tenant_id",
    "pack_id"
  ],
  "properties": {
    "tenant_id": {
      "type": "string"
    },
    "pack_id": {
      "type": "string"
    }
  }
}

Output Schema

{
  "$ref": "../../../../jsonschema/trust/evidence-pack/v1/schema.json"
}

get_receipt

MCP Tool

Fetch a signed checkout receipt by tenant and receipt identifier.

Input Schema

{
  "type": "object",
  "additionalProperties": false,
  "required": [
    "tenant_id",
    "receipt_id"
  ],
  "properties": {
    "tenant_id": {
      "type": "string"
    },
    "receipt_id": {
      "type": "string"
    }
  }
}

Output Schema

{
  "$ref": "../../../../jsonschema/trust/checkout-receipt/v1/schema.json"
}

lookup_merchant

MCP Tool

Lookup a business trust summary by tenant_id, domain, or platform account identifiers.

Input Schema

{
  "type": "object",
  "additionalProperties": false,
  "properties": {
    "tenant_id": {
      "type": "string"
    },
    "domain": {
      "type": "string"
    },
    "platform": {
      "type": "string"
    },
    "account_id": {
      "type": "string"
    }
  }
}

Output Schema

{
  "$ref": "../../../../jsonschema/trust/trust-registry-response/v1/schema.json"
}

verify_proof_bundle

MCP Tool

Verify a TrustProofBundle (signatures, anchors, and merkle proof where applicable).

Input Schema

{
  "type": "object",
  "additionalProperties": false,
  "required": [
    "proof_bundle"
  ],
  "properties": {
    "proof_bundle": {
      "$ref": "../../../../jsonschema/trust/trust-proof-bundle/v1/schema.json"
    }
  }
}

Output Schema

{
  "type": "object",
  "additionalProperties": false,
  "required": [
    "is_valid"
  ],
  "properties": {
    "is_valid": {
      "type": "boolean"
    },
    "errors": {
      "type": "array",
      "items": {
        "type": "string"
      }
    }
  }
}

Usage Example

Calling MCP tools from an AI agent

// MCP tool call request
{
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "name": "lookup_merchant",
    "arguments": {
      "domain": "example-store.com"
    }
  },
  "id": 1
}

// Response
{
  "jsonrpc": "2.0",
  "result": {
    "content": [{
      "type": "text",
      "text": "{\"trust_status\": \"verified\", \"tier\": \"enhanced\", ...}"
    }]
  },
  "id": 1
}