Skip to main content
MCP INTEGRATION

MCP Receipts

Generate a verifiable PEAC-Receipt for each Model Context Protocol (MCP) tool call, providing audit trails for AI agent operations.

MCP + PEAC Integration

The Model Context Protocol (MCP) is Anthropic’s open standard for connecting AI assistants to external tools and data sources. Originary extends MCP by generating a cryptographically signed receipt for each tool invocation.

This provides verifiable proof of which tools were called, when, by whom, and under what policy constraints - essential for compliance and auditing in production AI systems.

5-Minute Quickstart

  1. Install Originary MCP wrapper: npm install @originary/mcp-wrapper
  2. Wrap your existing MCP server with Originary middleware
  3. Configure policy constraints in peac.txt
  4. Each tool call now returns with a PEAC-Receipt header
  5. Query receipts via Originary dashboard for audit trails

Code Example

import { wrapMCPServer } from '@originary/mcp-wrapper'

const server = wrapMCPServer({
  name: 'my-tool-server',
  tools: [/* your MCP tools */],
  originary: {
    apiKey: process.env.ORIGINARY_API_KEY
  }
})

// Each tool call now includes PEAC-Receipt
// receipt proves: tool name, params, timestamp, policy

Add receipts to your MCP tools

Get audit trails for every AI tool invocation with Originary.

Get Started for $1