Serper MCP Server

A production-ready Model Context Protocol (MCP) server for Google Search via Serper.dev. This server supports both modern Streamable HTTP transport and legacy SSE (Server-Sent Events) transport, making it compatible with a wide range of MCP clients like Claude Desktop, Cursor, and llama.cpp.

Features

• Google Search Suite: Web, Images, Videos, News, Shopping, Scholar, and Patents search.
• Deep Search: A powerful serper_deep_search tool that combines search with automated scraping (similar to Tavily) to provide full page content instead of just snippets.
• Local Search: Specialized tools for Places and Maps.
• Visual Search: Google Lens reverse image search via URL.
• Web Scraper: Extract clean text content from any webpage.
• Dual Transport Support:
  • Streamable HTTP: Modern protocol used by Cursor, llama.cpp, etc.
  • SSE (Legacy): Standard protocol used by Claude Desktop.
• Dynamic API Key: Pass your Serper API key via query parameters for flexible deployment.

Deployment (Railway)

1. Fork/Clone this repository to GitHub.
2. Create a new project on Railway.app and connect it to your repository.
3. Railway will automatically detect the Procfile and deploy the server.
4. Once deployed, you will get a public URL (e.g., https://serper-mcp.up.railway.app).

Local Installation (Your PC)

If you prefer to run the server on your own machine:

1. Clone this repository.
2. Install dependencies:

   npm install
   

3. Build the project:

   npm run build
   

4. Start the server:

   npm start
   

The server will be active at http://localhost:3000/mcp?apiKey=YOUR_SERPER_API_KEY.

Connecting to Clients

Modern Clients (Cursor, llama.cpp, etc.)

Use the following URL:

https://your-domain.railway.app/mcp?apiKey=YOUR_SERPER_API_KEY

Legacy Clients (Claude Desktop)

Add this to your claude_desktop_config.json:

{
  "mcpServers": {
    "serper-remote": {
      "url": "https://your-domain.railway.app/sse?apiKey=YOUR_SERPER_API_KEY"
    }
  }
}

Available Tools

• serper_deep_search: (Recommended for Research) Searches Google and scrapes top results for full page content.
• serper_search: Standard Google search (titles and snippets).
• serper_images: Google Images search.
• serper_videos: Google Videos search.
• serper_places: Local business information.
• serper_maps: Location-based search.
• serper_news: Recent news articles.
• serper_shopping: Product and price search.
• serper_lens: Reverse image search.
• serper_scholar: Academic research search.
• serper_patents: Patent search.
• serper_autocomplete: Search suggestions.
• serper_webpage: Scrape text content from a URL.

License

MIT