Hey folks! If you’re like me, you love blending AI with everyday tasks — like ordering dinner without leaving your terminal. Zomato just dropped a shiny new Model Context Protocol (MCP) server on October 22, 2025, turning their food delivery magic into AI-friendly tools. This means you can discover restaurants, browse menus, build carts and place orders.
MCP? It’s an open standard (kicked off by Anthropic in late 2024) that lets AI agents like Gemini connect to external services seamlessly. Think of it as a universal plug for tools and APIs.
If you want to learn more about building an integrating an MCP server into Gemini CLI, check out tutorial below.
Medium
In this quick guide, we’ll integrate Zomato’s MCP server into Gemini CLI — Google’s open-source AI agent for your terminal (launched mid-2025). We’ll cover setup, config, and some fun interactions. Let’s get you ordering pizza with prompts in under 10 minutes!
What is the Zomato mcp-server-manifest?
Zomato’s mcp-server-manifest is an implementation of the Model Context Protocol (MCP) that exposes Zomato’s services as a tool for AI assistants. This allows you to conversationally interact with Zomato to discover restaurants, browse their menus, create a cart, and place an order. The server utilizes OAuth for secure authentication.
Prerequisites
Before we begin, ensure you have the following:
- Gemini CLI installed: If you haven’t already, you can install it by running the following command in your terminal:
npm install -g @google/gemini-cli
To learn more about Gemini CLI, check out the detailed tutorial series given below.
- A Zomato Account: You will need your Zomato account credentials to authenticate and place orders.
Integrating Zomato with Gemini CLI: A Step-by-Step Guide
The key to this integration lies in configuring the Zomato MCP server within the Gemini CLI’s settings. Here’s how to do it:
Step 1: Locate your Gemini CLI settings file
The Gemini CLI configuration is stored in a JSON file located in your home directory. You can find it at ~/.gemini/settings.json.
code ~/.gemini
Step 2: Add the Zomato MCP Server Configuration
Open the settings.json file in your favorite text editor. You will need to add a new entry to the mcpServers object. This object contains the configurations for all the external tools you want to integrate with Gemini CLI.
"mcpServers": {
"zomato-mcp": {
"command": "npx",
"args": [
"mcp-remote",
"https://mcp-server.zomato.com/mcp"
]
}
},
Check out the official Github repository for the MCP server.
Zomato MCP Server
An mcp server for your food ordering needs.
Disclaimer : This is only for testing purposes and Zomato disclaims any and all liabilities that may arise due to erroneous / non-functionality of the MCP integration.
Supported Features
- 🔎 Restaurant Discovery - Find nearby restaurants based on your location and preferences.
- 📒 Menu Browsing - Browse through detailed menus with prices, descriptions, and ratings.
- 🛒 Cart Creation - Add items to your cart and customize orders with ease.
- 🥗 Food Ordering - Place orders seamlessly with order tracking support.
- 💳 QR code payment - Complete secure payments using QR code integration.
Installation Guide
⚠️ OAuth Redirect URI Warning: Currently, we have only whitelisted the following redirect URIs for OAuth authentication. Please reach out to us to enable your client:
claude://claude.ai/settings/connectorshttps://chatgpt.com/connector_platform_oauth_redirecthttp://localhosthttp://127.0.0.1https://claude.ai/api/mcp/auth_callbackhttps://insiders.vscode.dev/redirecthttps://oauth.pstmn.io/v1/callbackhttps://vscode.dev/redirect
Install in VsCode
One Click Installation
Manual Installation
Add this…
Explanation of the configuration:
zomato: This is a user-defined name for the MCP server. You can choose any name you like.
command: npx **: This specifies the command to execute to start the MCP server. In this case, we are using npx, which is a package runner tool that comes with npm.
**args: […]: This is an array of arguments to be passed to the command.
mcp-server: This is the name of the package that will run the MCP server.
https://github.com/Zomato/mcp-server-manifest: This is the URL to the Zomato mcp-server-manifest on GitHub.
Step 3: Save and Restart Gemini CLI
Save the settings.json file and restart your Gemini CLI. If it’s already running, you can exit and relaunch it.
Step 4: Verify the Integration
Once Gemini CLI restarts, it will automatically connect to the Zomato MCP server. The first time you try to access it requires authentication , you will be prompted to log in to your Zomato account through a secure OAuth flow. Once done, you will be redirected back to your terminal.
To verify that the integration is successful, you can use the /mcp command in the Gemini CLI:
/mcp
This command will list all the configured MCP servers. You should see zomato in the list of connected servers.
Interacting with Zomato through Gemini CLI
Now for the fun part! You can start interacting with Zomato using natural language queries directly in your Gemini CLI. Here are a few examples to get you started:
- To find restaurants: Find me some biryani restaurants near me.
- To see the menu of a restaurant: Show me the menu for Restaurant Name.
- To add items to your cart: Add one chicken biryani to my cart.
- To view your cart: Show my cart.
- To place an order: Place this order.
Check out the demo given below for verification and interaction with the Zomato MCP Server using Gemini CLI.
A Powerful Combination for a Seamless Workflow
The integration of the Zomato mcp-server with Gemini CLI is a fantastic example of how the command line is evolving into a more interactive and powerful tool. This seamless connection allows developers and command-line enthusiasts to stay within their preferred environment while accessing essential real-world services. So, the next time you’re deep in your coding workflow and hunger strikes, you know you’re just a command away from your next meal.
Top comments (0)