Zomato -Demo

The Zomato MCP Server enables Claude Desktop (or any MCP-compatible client) to interact with a live food ordering workflow through chat. Once configured, you can look up restaurants by location and preferences, browse detailed menus with prices and categories, add items to a cart, and run through a simulated checkout and payment flow. The server exposes conversational commands that map to real-time actions such as restaurant discovery, menu retrieval, cart management, and simulated ordering. For authentication, the server currently whitelists localhost and chatgpt.com, so ensure your client runs from an allowed origin when testing OAuth flows.

To use it, connect Claude Desktop to the MCP server by adding the zomato-mcp entry in claude_desktop_config.json as shown in the setup guide. After saving the configuration and restarting Claude Desktop, you can start a chat and issue natural-language requests like “Show me Italian restaurants near me,” “What’s on the pizza menu,” or “Add Margherita to my cart.” The server translates these prompts into MCP API calls and returns structured results (restaurant details, menus, cart items, and simulated payments) that you can reference in the chat.

If you’re testing interactions, use the provided demo videos as a reference for typical user flows—from discovering restaurants to simulating a payment, ensuring you understand how the MCP endpoints respond during each step of the conversation.

Prerequisites:

• Node.js installed (verify with node -v)
• Basic understanding of JSON configuration and MCP concepts
• Claude Desktop or another MCP-compatible client to connect to the server

Step 1: Install Node.js (if needed)

• Download and install from https://nodejs.org

Step 2: Prepare the MCP configuration (on the client)

• Open Claude Desktop settings: Settings → Developer → Edit Config
• Add the Zomato MCP Server configuration shown below to claude_desktop_config.json:

{
  "mcpServers": {
    "zomato-mcp": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://mcp-server.zomato.com/mcp"
      ]
    }
  }
}

Step 3: Save and restart Claude Desktop

• Save the file and restart Claude Desktop to load the MCP server.

Step 4: Verify the connection

• In Claude Desktop, select the Zomato MCP server and start a chat to test restaurant lookups and menu retrieval.

Tips and known considerations:

• The MCP server relies on npx to fetch and run the mcp-remote client from the npm registry. Ensure you have network access and npm cache available.
• OAuth with Zomato currently whitelists localhost and chatgpt.com; if testing from another origin, you may need to adjust allowed origins or use a supported environment.
• If you encounter connectivity or authentication issues, double-check the MCP server URL and ensure the CLAUDE Desktop config JSON is valid JSON (no trailing commas).
• The configuration shown uses the canonical remote URL https://mcp-server.zomato.com/mcp; replace with a different MCP endpoint only if you are instructed to do so by the server provider.
• This MCP setup is intended for demo and testing scenarios and may not reflect production-grade authentication or payment flows.