OpenRouter MCP Server

A Model Context Protocol (MCP) server that provides integration with OpenRouter.ai, allowing access to various AI models through a unified interface.

Features

• Access to all OpenRouter.ai models through MCP tools
• Advanced model search and filtering capabilities
• Automatic rate limiting and error handling
• Model information caching for optimal performance
• Comprehensive model capability tracking

Installation

npm install @mcpservers/openrouterai

Configuration

Add the server to your MCP settings configuration file (cline_mcp_settings.json or claude_desktop_config.json):

{
  "mcpServers": {
    "openrouterai": {
      "command": "npx",
      "args": ["@mcpservers/openrouterai"],
      "env": {
        "OPENROUTER_API_KEY": "your-api-key-here",
        "OPENROUTER_DEFAULT_MODEL": "optional-default-model"
      }
    }
  }
}

Required environment variables:

• OPENROUTER_API_KEY: Your OpenRouter API key (get one at https://openrouter.ai/keys)
• OPENROUTER_DEFAULT_MODEL (optional): Default model to use for chat completions

Available Tools

chat_completion

Send messages to OpenRouter.ai models:

{
  model?: string;          // Optional if default model is set
  messages: {
    role: "system" | "user" | "assistant";
    content: string;
  }[];
  temperature?: number;    // Optional, defaults to 1.0
}

search_models

Search and filter available models:

{
  query?: string;          // Search in name/description
  provider?: string;       // Filter by provider
  minContextLength?: number;
  maxContextLength?: number;
  maxPromptPrice?: number;
  maxCompletionPrice?: number;
  capabilities?: {
    functions?: boolean;
    tools?: boolean;
    vision?: boolean;
    json_mode?: boolean;
  };
  limit?: number;          // Default: 10, max: 50
}

get_model_info

Get detailed information about a specific model:

{
  model: string;           // Model identifier
}

validate_model

Check if a model ID is valid:

{
  model: string;           // Model identifier to validate
}

For detailed documentation, development setup, and implementation details, see CONTRIBUTING.md.