Instacart MCP Server
Enables AI agents to search for products, manage shopping carts, and place grocery orders on Instacart using browser automation. It includes comprehensive tools for store discovery, product searching, and secure checkout with explicit user confirmation.
README
@striderlabs/mcp-instacart
MCP server that gives AI agents the ability to search products, manage cart, and place grocery orders on Instacart.
Built by Strider Labs ā building the action layer for AI agents.
Features
- š Search Products ā Find products across Instacart with prices and availability
- š Cart Management ā Add items, view cart, clear cart
- š¦ Place Orders ā Preview and place orders with delivery time selection
- š Session Persistence ā Cookies saved locally for seamless re-authentication
- š”ļø Order Safety ā Requires explicit confirmation before placing orders
Installation
npm install @striderlabs/mcp-instacart
Or install globally:
npm install -g @striderlabs/mcp-instacart
Prerequisites
This package requires Playwright browsers. Install them with:
npx playwright install chromium
MCP Client Configuration
Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"instacart": {
"command": "npx",
"args": ["@striderlabs/mcp-instacart"]
}
}
}
Other MCP Clients
npx @striderlabs/mcp-instacart
Available Tools
Authentication
| Tool | Description |
|---|---|
instacart_status |
Check login status and session info |
instacart_login |
Initiate login flow (returns URL for manual login) |
instacart_logout |
Clear saved session and cookies |
Shopping
| Tool | Description |
|---|---|
instacart_search |
Search for products by name |
instacart_stores |
List available stores for delivery location |
instacart_set_address |
Set delivery address or zip code |
Cart
| Tool | Description |
|---|---|
instacart_add_to_cart |
Add a product to cart |
instacart_view_cart |
View cart contents and totals |
instacart_clear_cart |
Remove all items from cart |
Orders
| Tool | Description |
|---|---|
instacart_preview_order |
Preview order before placing |
instacart_place_order |
Place order (requires confirm=true) |
Usage Examples
Check Session Status
User: Am I logged into Instacart?
Agent: [calls instacart_status]
ā Returns login state, email, saved address
Search for Products
User: Find organic bananas on Instacart
Agent: [calls instacart_search with query="organic bananas"]
ā Returns list of products with prices
Build a Shopping List
User: Add milk, eggs, and bread to my Instacart cart
Agent:
[calls instacart_add_to_cart with product="milk"]
[calls instacart_add_to_cart with product="eggs"]
[calls instacart_add_to_cart with product="bread"]
ā Items added to cart
View Cart
User: What's in my Instacart cart?
Agent: [calls instacart_view_cart]
ā Returns items, quantities, subtotal, fees, total
Place Order
User: Place my Instacart order
Agent: [calls instacart_place_order with confirm=false]
ā Returns preview asking for confirmation
User: Yes, place it
Agent: [calls instacart_place_order with confirm=true]
ā Order placed, returns confirmation
Authentication Flow
Instacart requires browser-based login. The flow:
- Call
instacart_loginā returns login URL - User opens URL and logs in manually
- Cookies are automatically saved to
~/.strider/instacart/ - Future sessions use saved cookies
Session cookies persist until they expire or instacart_logout is called.
Configuration
Session data is stored in:
~/.strider/instacart/
āāā cookies.json # Browser cookies
āāā session.json # Session metadata
Order Safety
The instacart_place_order tool has a built-in safety mechanism:
- Without
confirm=true: Returns a preview, does not place order - With
confirm=true: Actually places the order
Agents should ALWAYS show the preview and get explicit user confirmation before calling with confirm=true.
Technical Details
- Uses Playwright for browser automation
- Runs headless by default
- Includes stealth patches to avoid detection
- Supports all Instacart stores available in your area
Limitations
- Requires manual login (no programmatic auth)
- Browser automation may break if Instacart updates their UI
- Some anti-bot measures may require solving CAPTCHAs manually
- Payment methods must be pre-configured in your Instacart account
Troubleshooting
"Not logged in" errors
Run instacart_login and complete the login flow manually.
Timeout errors
Instacart pages may be slow. The server will retry automatically, but you can also:
- Check your internet connection
- Try again after a moment
"Could not find element" errors
Instacart may have updated their UI. Please file an issue with details.
CAPTCHA challenges
Some actions may trigger CAPTCHA. You'll need to:
- Open Instacart in a regular browser
- Complete the CAPTCHA
- Try the MCP action again
Development
# Clone the repo
git clone https://github.com/markswendsen-code/mcp-instacart
cd mcp-instacart
# Install dependencies
npm install
# Install Playwright browsers
npx playwright install chromium
# Build
npm run build
# Run locally
node dist/index.js
License
MIT
Links
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.