Milkbasket MCP Server
Enables AI assistants to interact with Milkbasket for grocery shopping, including OTP authentication, product search, cart management, wallet balance checks, and order summaries through natural language.
README
@anujsup/milkbasket-mcp
AI-powered grocery shopping for India — a Model Context Protocol (MCP) server for the Milkbasket API.
Connect your AI assistant (Cursor, Claude Desktop, Windsurf, or any MCP client) to your Milkbasket account and let it handle your daily groceries — log in with OTP, browse products, build your cart, check your wallet, and review your full order summary, all through natural language.
What you can do with this:
- Ask your AI to find products — "Search for low-fat milk under ₹50"
- Build and manage your cart — "Add 2 Amul butter, remove the lassi"
- Check your order before delivery — "Show me my cart with the full bill"
- Keep track of your wallet — "What's my Milkbasket balance?"
- Fetch account details — "What city and hub am I assigned to?"
Prerequisites
- Node.js 18 or higher
- A Milkbasket account (Indian mobile number)
- An MCP client (Cursor, Claude Desktop, etc.)
Installation
Option 1: npx (recommended, no install needed)
{
"mcpServers": {
"milkbasket": {
"command": "npx",
"args": ["-y", "@anujsup/milkbasket-mcp@latest"]
}
}
}
Option 2: Global install
npm install -g @anujsup/milkbasket-mcp
{
"mcpServers": {
"milkbasket": {
"command": "milkbasket-mcp"
}
}
}
Option 3: Local install
npm install @anujsup/milkbasket-mcp
{
"mcpServers": {
"milkbasket": {
"command": "node",
"args": ["node_modules/@anujsup/milkbasket-mcp/dist/index.js"]
}
}
}
Cursor setup
Add the config to .cursor/mcp.json in your project root (project-level) or ~/.cursor/mcp.json (global):
{
"mcpServers": {
"milkbasket": {
"command": "npx",
"args": ["-y", "@anujsup/milkbasket-mcp@latest"]
}
}
}
Reload Cursor after saving.
Available Tools
Authentication
| Tool | Description |
|---|---|
auth_request_otp |
Send OTP to your registered mobile number |
auth_verify_otp |
Verify OTP and log in |
auth_status |
Check if you are currently authenticated |
auth_logout |
Log out and clear stored tokens |
User
| Tool | Description |
|---|---|
get_user_details |
Fetch your name, email, city, hub, and membership info |
get_wallet_balance |
Fetch your current Milkbasket wallet balance and cashback |
Products
| Tool | Description |
|---|---|
list_products |
List featured/flash-deal products (paginated) |
search_products |
Search products by name or keyword (paginated) |
get_hero_collections |
Fetch hero banner collections shown on the home screen |
Cart
| Tool | Description |
|---|---|
add_to_cart |
Add a new product or increase quantity in the cart |
remove_from_cart |
Decrease quantity or remove a product from the cart |
get_basket |
Lightweight cart snapshot — item count, product IDs, basic totals |
get_extended_basket |
Full cart view — product names, images, discounts, delivery fee, GST, final payable amount |
Authentication Flow
1. auth_request_otp → { phone: "9XXXXXXXXX" }
2. Receive OTP via SMS
3. auth_verify_otp → { phone: "9XXXXXXXXX", otp: "XXXX" }
4. You are now logged in
Security & Privacy
This server runs 100% locally on your machine. Your phone number, OTP, and tokens never leave your device — they are never sent to any third party, including the AI assistant. Here's exactly what happens:
- Your phone number is sent directly from your machine to Milkbasket's servers to request an OTP
- The OTP you share with the AI is passed directly from your machine to Milkbasket to verify login
- The AI model only sees the response (success/failure) — it cannot intercept or store your credentials
- Access tokens are stored in your OS keychain (macOS Keychain, Windows Credential Manager, Linux Secret Service) using
keytar— the same secure storage used by apps like VS Code and 1Password - If
keytaris unavailable, an AES-256-GCM encrypted file is used as fallback - You can log out at any time using
auth_logout, which clears all stored tokens immediately
Cart Flow
search_products → find productId and price
add_to_cart → { productId, price, quantity }
remove_from_cart → { productId, price, quantity: 0 } ← set 0 to delete
get_extended_basket → full checkout view
What you can ask your AI
Once connected, just talk to your AI naturally:
| What you say | What happens |
|---|---|
| "Log me in to Milkbasket" | Sends OTP → verifies → stores token |
| "Search for paneer" | Calls search_products and returns matches |
| "Add 2 Amul Gold milk to my cart" | Calls add_to_cart with the right product |
| "Remove the lassi from my cart" | Calls remove_from_cart with quantity 0 |
| "Show me my cart with full bill" | Calls get_extended_basket with itemized bill |
| "What is my wallet balance?" | Calls get_wallet_balance |
| "What are my account details?" | Calls get_user_details — name, city, hub, membership |
| "What deals are on today?" | Calls list_products for flash deals |
| "Log me out" | Clears tokens from keychain |
Example usage in Cursor
Search for Amul milk products
→ search_products({ query: "amul milk" })
Add 2 units to cart
→ add_to_cart({ productId: 1234, price: 28, quantity: 2 })
View full cart with bill breakdown (product names, discounts, delivery fee, GST, payable)
→ get_extended_basket()
Reduce quantity by 1
→ remove_from_cart({ productId: 1234, price: 28, quantity: 1 })
Remove product entirely
→ remove_from_cart({ productId: 1234, price: 28, quantity: 0 })
Check wallet balance
→ get_wallet_balance()
Get account info
→ get_user_details()
License
MIT © Anujsup
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.