> For the complete documentation index, see [llms.txt](https://rwapact.gitbook.io/whitepaper/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://rwapact.gitbook.io/whitepaper/integrations-and-standards/mcp.md).

# MCP

## MCP middleware integration

Model Context Protocol, or MCP, gives agents a structured path to request tool actions. A generated tool call is not an execution authorization.

RWAPACT places middleware between the agent and the execution MCP. The middleware intercepts the request, runs the Pact, then forwards only an authorized request.

```
Agent → MCP middleware → RWAPACT Pact → Execution MCP
```

This separation keeps intent and authorization distinct. An agent may request an action through an MCP workflow. The request reaches execution only after the Pact evaluates it.

### Pact evaluation

The middleware evaluates the request through the five Pact layers:

* **KRO** evaluates RWA-specific risk.
* **KSO** checks that the agent session remains active and in scope.
* **KPV** checks the requested asset and amount against policy.
* **KGR** evaluates gas and the execution path.
* **KAR** records the resulting authorization or rejection.

If all required conditions pass, the middleware can forward the request to the execution MCP. If a required condition fails, it rejects the request before forwarding. KAR records that decision.

### Example workflow

An autonomous agent requests a tokenized stock trade through an MCP workflow. The middleware receives the tool-call request before the execution MCP.

KRO assesses the trade risk. KSO validates the current session. KPV evaluates the requested asset and amount. KGR evaluates the execution conditions.

The execution MCP receives the request only when the required Pact conditions pass. MCP remains the agent interaction and execution interface. RWAPACT remains the decision layer before that execution path.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://rwapact.gitbook.io/whitepaper/integrations-and-standards/mcp.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
