MCP ExplorerExplorer

Rest To Mcp

@viktormarinhoon a year ago
6 MIT
FreeCommunity
AI Systems
A bridge that enables LLMs to access and interact with documented REST APIs through the Model Context Protocol (MCP)

Overview

What is Rest To Mcp

rest-to-mcp is a bridge that enables large language models (LLMs) to access and interact with documented REST APIs through the Model Context Protocol (MCP).

Use cases

Use cases include integrating LLMs with third-party services, automating API interactions, and enhancing applications with dynamic data retrieval from REST APIs.

How to use

To use rest-to-mcp, install Deno, run the MCP server, and connect your LLM to discover available API endpoints and make HTTP requests.

Key features

Key features include automatic discovery of API documentation, support for various HTTP methods (GET, POST, PUT, DELETE), and the ability to handle JSON request and response bodies.

Where to use

rest-to-mcp can be used in fields such as software development, data integration, and any application requiring interaction with REST APIs.

Content

rest-to-mcp

A bridge that enables any LLM to access and interact with documented REST APIs through the Model Context Protocol (MCP).

Overview

rest-to-mcp provides a seamless adapter layer that allows language models to:

  1. Discover available endpoints from REST APIs that publish OpenAPI/Swagger documentation
  2. Make direct HTTP requests to these endpoints

This enables LLMs to interact with external REST services without requiring custom integration code for each API.

How It Works

This project implements an MCP server with two primary tools:

1. discover-endpoints

Automatically locates API documentation by probing common paths where OpenAPI/Swagger specs are typically published:

  • /openapi.json
  • /swagger.json
  • /swagger/v1/swagger.json
  • /api-docs
  • /api-docs.json

2. make-request

Allows direct HTTP interactions with any REST endpoint:

  • Supports GET, POST, PUT, DELETE methods
  • Handles JSON request and response bodies
  • Returns formatted JSON responses to the LLM

Developing

deno install

Start the MCP server:

deno run dev

This launches the server using the MCP Inspector for testing and debugging.

Integration

LLMs can connect to this server through the Model Context Protocol, enabling them to:

  1. First discover what endpoints are available on a REST API
  2. Make appropriate requests to those endpoints
  3. Process the JSON responses

License

Project licensed under the MIT License.

More information on the LICENSE file.

Tools

No tools

Comments

Recommend MCP Servers

View All MCP Servers