Content
# Meta API MCP Server
A meta API Gateway server that works with the Model Context Protocol (MCP). You can connect any API to LLMs (Claude, GPT, etc.) through MCP. This enables AI assistants to interact directly with APIs and access real-world data sources.
## Features
- 🔄 Multi-API support: Manage multiple APIs through a single server
- 🛠️ Easily add APIs with JSON configuration files
- 📋 Automatically convert Postman Collections to MCP tools
- 🔌 Comprehensive support for HTTP APIs (GET, POST, PUT, DELETE, PATCH)
- 🔒 Various authentication methods (API Key, Bearer Token)
- 📁 Load configurations from local files or remote URLs
- 📑 Support for configuration file lists
## API Editor Tool
A user-friendly editor tool has been developed to create or edit JSON configuration files:
[MCP API Editor](https://savhascelik.github.io/mcp-api-editor/)
## Installation
### Global Installation with NPM (Recommended)
```bash
npm install -g meta-api-mcp-server
```
### Installation from Source Code
```bash
git clone https://github.com/savhascelik/meta-api-mcp-server.git
cd meta-api-mcp-server
npm install
```
## Usage
### As a Command Line Tool
```bash
# Load from default api-configs/ folder
meta-api-mcp-server
# Specify a configuration file (in the directory where you run the server, there should be a folder with this name and structured json in it)
meta-api-mcp-server path/to/config.json
# Load from a specific folder
meta-api-mcp-server path/to/configs/
# Load from a remote URL
meta-api-mcp-server https://example.com/api-config.json
# Load from a remote configuration list
meta-api-mcp-server https://example.com/config-list.json
# Load from a Postman Collection ( your filename must contain the word ‘postman’, I'll bind it to a variable when I have time )
meta-api-mcp-server path/to/My-API.postman_collection.json
```
### Using with Cursor or Other MCP Clients
To connect to an MCP client like Cursor, configure your `mcp.json` file as follows:
```json
{
"mcpServers": {
"myApiServer": {
"command": "meta-api-mcp-server",
"args": [
],
"env": {
"MCP_CONFIG_SOURCE":"api-configs/flexweather-endpoints.json",
"API_KEY": "your-api-key-here"
}
}
}
}
```
```json
{
"mcpServers": {
"myApiServer": {
"command": "meta-api-mcp-server",
"args": [
"server.js",
"path/to/api-config.json"
],
"env": {
"EXAMPLE_API_KEY": "your-api-key-here"
}
}
}
}
```
```json
{
"mcpServers": {
"flexweather": {
"command": "node",
"args": [
"server.js"
],
"env": {
"MCP_CONFIG_SOURCE":"api-configs/flexweather-endpoints.json"
}
}
}
}
```
```json
{
"mcpServers": {
"lemonsqueezy": {
"command": "node",
"args": [
"server.js"
],
"env": {
"MCP_CONFIG_SOURCE":"api-configs/lemon-squeezy-api.json",
"LEMON_SQUEEZY_API_KEY": ""
}
}
}
}
```
## Postman Collection Conversion
Using your existing Postman collections with Meta API MCP Server is now very easy! You can use hundreds of ready-made APIs without writing a single line of code.
1. Export your Postman collection (in v2.1.0 format)
2. Start Meta API MCP Server with the collection file:
```bash
meta-mcp my-collection.postman_collection.json
```
3. The server will automatically:
- Analyze all endpoints
- Detect the authentication method
- Extract path/query parameters
- Analyze request body structure
- Create MCP tools
4. Add your API key to the `.env` file (the server will tell you which environment variable to use)
### Supported Postman Collection Features
- ✅ Multi-level folder structure
- ✅ Bearer token authentication
- ✅ API Key authentication
- ✅ Path parameters
- ✅ Query parameters
- ✅ Headers
- ✅ JSON request body
- ✅ Postman variables (like {{api_url}})
## API Editor Tool
A user-friendly editor tool has been developed to create or edit JSON configuration files:
[MCP API Editor](https://savhascelik.github.io/mcp-api-editor/)
With this web tool, you can:
- Create API configurations through a visual interface
- Edit existing JSON configurations
- Convert Postman collections to MCP-compatible configuration files
- Validate configuration files
- Export your configurations as JSON
**Postman Collections:** You can upload your existing Postman collections to the editor tool and automatically convert them to MCP-compatible configurations. This allows you to quickly use your existing collections instead of configuring APIs from scratch.
The editor makes it easy to manage tool names, parameters, and all other configuration options.
## Project Structure
The codebase is organized in a modular way to facilitate maintenance and extension:
```
meta-api-mcp-server/
├── serve.js
├── api-configs/ # Default config directory
└── package.json
```
## API Configuration File Format
You can manually configure APIs using the following JSON format:
```json
{
"apiId": "my-api",
"handlerType": "httpApi",
"baseUrl": "https://api.example.com",
"authentication": {
"type": "bearerToken",
"envVariable": "MY_API_TOKEN"
},
"endpoints": [
{
"mcpOperationId": "getUsers",
"description": "Get a list of users",
"targetPath": "/users",
"targetMethod": "GET",
"parameters": [
{
"name": "page",
"in": "query",
"required": false,
"type": "integer",
"description": "Page number"
}
]
}
]
}
```
## License
MIT
