meta-api

Meta API MCP Server is a gateway that connects multiple APIs to the Model Context Protocol (MCP), enabling AI assistants to interact with real-world data sources through a single, unified interface. The server loads API configurations from local files or remote URLs and automatically exposes MCP tools for each configured endpoint. This makes it easy to manage multiple APIs, support common HTTP methods (GET, POST, PUT, DELETE, PATCH), and handle various authentication schemes such as API keys and Bearer tokens. You can also convert Postman collections into MCP-compatible configurations, expanding the range of APIs you can integrate without manual coding. The built-in API editor helps you create, edit, and validate JSON configurations, and you can point MCP to an editor-generated configuration source via the MCP_CONFIG_SOURCE environment variable or by passing a config path to the CLI.

To use Meta API MCP Server with a client like Cursor, you define an mcp.json that references your server and its configuration source. The server then exposes individual MCP tools (one per API or endpoint) that you can invoke through MCP clients, enabling your AI assistants to call those APIs as needed. This workflow is ideal for building AI-assisted integrations, dashboards, or agents that rely on up-to-date external data from multiple services.

Prerequisites:

• Node.js (LTS) and npm installed on your system
• Git for cloning repositories (optional if using npm install globally)

Installation steps:

1. Global installation (recommended for quick start): npm install -g meta-api-mcp-server

2. Install from source (development-ready): git clone https://github.com/savhascelik/meta-api-mcp-server.git cd meta-api-mcp-server npm install

3. Run the server:

   • If installed globally: meta-api-mcp-server
   • If running from source: node serve.js

4. Optional: prepare an API configuration file and specify its path when starting the server, for example: node serve.js path/to/api-config.json

Note: The project structure exposes a default api-configs/ folder. You can populate it with your configuration files or point the server to a remote URL that returns configuration data.

Tips and common considerations:

• Environment variables: use MCP_CONFIG_SOURCE to point to a local path or URL that returns your API configurations. You can also use authentication-related environment variables as required by each API (e.g., API keys or tokens).
• Postman conversion: leverage the Postman collection conversion feature to generate MCP-compatible configurations automatically.
• Multiple APIs: you can manage many APIs under a single MCP server instance. Each API or tool can be exposed as a separate MCP tool with its own endpoint and authentication configuration.
• Debugging: if a tool isn’t appearing in MCP clients, verify the config file structure, ensure the server has access to the config source, and check environment variables for correct values.
• Security: avoid hard-coding credentials in configuration files. Use environment variables and secret management practices where possible.