Model Context Protocol (MCP)
Verifiable receipts for LLM tool calls and context sharing
Overview
Model Context Protocol (MCP) is an open standard for connecting LLMs to external data sources and tools. Originary extends MCP with PEAC-Receipt verification, providing cryptographic proof of tool invocations and policy compliance.
When an LLM calls an MCP tool that requires payment or policy adherence, Originary issues a signed receipt that can be verified by downstream systems. This enables trusted tool calling workflows with full audit trails.
Use Cases
Paid Tool Access
Monetize MCP tools by requiring payment before execution. PEAC-Receipts prove payment completion for downstream verification.
Multi-Tool Workflows
Chain multiple MCP tool calls together with verifiable receipts showing the complete workflow execution path.
Policy Enforcement
Verify that LLMs have obtained necessary permissions before accessing sensitive MCP resources.
Implementation
MCP Server with PEAC-Receipt
{
"jsonrpc": "2.0",
"id": 1,
"method": "tools/call",
"params": {
"name": "fetch_data",
"arguments": {
"url": "https://api.example.com/data"
}
},
"receipt": "eyJhbGc...PEAC-Receipt-signature"
}MCP servers can verify attached receipts before executing tools, ensuring proper authorization and payment.
Benefits for MCP Providers
- Monetize MCP tools with pay-per-call pricing
- Cryptographic verification of tool invocations
- Complete audit trail of LLM tool usage
- Policy enforcement at the tool level
- Compatible with standard MCP implementations