How MCP.Link Works

Seamlessly integrate any API with AI agents through automatic conversion to the Model-Context-Protocol standard.

Overview

MCP.Link automatically converts traditional RESTful APIs into interfaces compliant with the MCP standard, enabling any Web API to seamlessly integrate with the AI Agent ecosystem without modifying the original API implementation.

  AI Agents                                 MCP Link                              Traditional API
+-------------+                        +------------------+                     +----------------+
|             |                        |                  |                     |                |
| LLM-based   |                        | MCP-compatible   |                     | RESTful        |
| Autonomous  |  Agent requests        | endpoints        |  OpenAPI Description| SOAP (TODO)    |
| Agents      +----------------------->+ ↓                +-------------------->+ GraphQL (TODO) |
|             |                        | Auto-conversion  |                     | RPC (TODO)     |
|             |                        | ↓                |                     |                |
+-------------+                        | OpenAPI Schema   |                     +----------------+

Why MCP.Link?

There is a notable gap in the current AI Agent ecosystem:

Most MCP Servers are simple wrappers around Web APIs
Functionality interfaces may not be complete, depending on developer implementation
Manual creation of MCP interfaces is time-consuming and error-prone
Lack of standardized conversion processes

MCP.Link solves these issues through automation and standardization, allowing any API to easily join the AI-driven application ecosystem.

Key Features

Automatic Conversion: Generate complete MCP Servers based on OpenAPI Schema
Seamless Integration: Make existing RESTful APIs immediately compatible with AI Agent calling standards
Complete Functionality: Ensure all API endpoints and features are correctly mapped
Zero Code Modification: Obtain MCP compatibility without modifying the original API implementation
Open Standard: Follow the MCP specification to ensure compatibility with various AI Agent frameworks

How It Works

1. OpenAPI Specification

The process starts with an OpenAPI specification, which is a standard format for describing RESTful APIs. This specification contains information about the API's endpoints, parameters, request/response formats, and more.

2. MCP Adapter

The MCP Adapter reads the OpenAPI specification and dynamically creates MCP tools that correspond to the API's endpoints. It handles the translation between the AI assistant's requests and the API's expected format.

You can run the adapter locally on your machine (localhost:3000) or use our hosted version on onrender.com.

3. AI Assistant Integration

AI assistants can connect to the MCP Adapter using the MCP Link, which contains all the information needed to establish the connection. The AI assistant can then use the API's functionality as if it were a native tool.

4. API Backend

The MCP Adapter forwards requests to the actual API backend, which processes the requests and returns the results. The adapter then translates these results back into a format that the AI assistant can understand.

MCP Link Format

The MCP Link is a URL-like string that contains all the information needed to connect to an API through the MCP Adapter. There are two encoding formats available:

Default Format (URL Encoded)

https://[SERVER_URL]/sse?s=[SPEC_URL]&u=[BASE_URL]&h=[KEY:VALUE]

Base64 Format (JSON Encoded)

https://[SERVER_URL]/sse?code=[BASE64_ENCODED_JSON]

Parameters

SERVER_URL: Either localhost:3000 (local) or https://mcp-openapi-to-mcp-adapter.onrender.com (hosted)
s: URL to the OpenAPI specification (YAML/JSON)
Example: https://api.example.com/spec.yaml
u: Base URL of the target service where API requests will be forwarded
Example: https://api.example.com
h: (Optional) HTTP headers to include in requests to the target service
Example: Authorization:Bearer token
You can include multiple headers by adding multiple h= parameters
code: (Alternative format) Base64 encoded JSON containing all parameters
Example: {"s":"https://api.example.com/spec.yaml","u":"https://api.example.com","h":{"Authorization":"Bearer token"}} (base64 encoded)

Usage in AI Agents

AI agents can use MCP Links in their configuration to access API functionality:

{
  "mcpServers": {
    "@service-name": {
      "url": "http://localhost:8080/sse?s=[OpenAPI-Spec-URL]&u=[API-Base-URL]&h=[Auth-Header]:[Value-Prefix]"
    }
  }
}

These URLs allow any API with an OpenAPI specification to be immediately converted into an MCP-compatible interface accessible to AI Agents.

Ready to try it out? Browse our pre-configured MCP links or create your own.

Browse MCP Links Create Custom Link