Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Cart MCP Server search for trending products in electronics"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP (Model Context Protocol) server for the Cart e-commerce intelligence API. This server exposes Cart's full API as tools that AI agents can call directly.
Tool | Description |
| Search for e-commerce stores by keyword, platform, language, traffic |
| Get full store details by domain |
| List products for a store |
| Get traffic analytics for a store |
| Get the technology stack for a store |
| Get ads detected for a store |
| Compare 2-10 stores side by side |
| Search for products across all tracked stores |
| Get a single product by ID |
| Get trending products and growing stores |
| Search for e-commerce advertisements |
| Get a single ad by ID |
| Search for product suppliers and vendors |
| Get a market overview for a niche keyword |
| Get API key info and usage stats |
Related MCP server: SellerChamp MCP Server
Node.js 18 or later
A Cart API key (get one at usecart.com)
Add this to your Claude Desktop configuration file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"cart": {
"command": "npx",
"args": ["@usecart/mcp-server"],
"env": {
"CART_API_KEY": "cart_sk_..."
}
}
}
}Add this to your VS Code MCP settings (.vscode/mcp.json or user settings):
{
"mcpServers": {
"cart": {
"command": "npx",
"args": ["@usecart/mcp-server"],
"env": {
"CART_API_KEY": "cart_sk_..."
}
}
}
}claude mcp add cart -- npx @usecart/mcp-serverSet the environment variable before running:
export CART_API_KEY=cart_sk_...Any MCP-compatible client can use this server. The configuration is always:
Command: npx
Args: ["@usecart/mcp-server"]
Environment: CART_API_KEY must be set
Variable | Required | Description |
| Yes | Your Cart API key (starts with |
| No | Override the API base URL (defaults to |
# Install dependencies
npm install
# Build
npm run build
# Run locally
CART_API_KEY=cart_sk_... node dist/index.jsMIT
Compare 2 to 10 e-commerce stores side by side. Returns full store profiles for each domain so you can compare traffic, products, platforms, and more.
Inputs: domains (required).
Get information about the authenticated Cart API account, including the current plan, email, daily request count, and request limit. Use this to check remaining API quota.
Inputs: no named parameters.
Get full details for a single advertisement by its Cart ad ID.
Inputs: id (required).
Get a market overview for a niche keyword. Returns the total number of stores and products in the niche, average price, top stores, and trending products. Great for market research and competitive analysis.
Inputs: keyword (required).
Get full details for a single product by its Cart product ID. Returns the product title, price, image, vendor, store domain, and timestamps.
Inputs: id (required).
Get the full profile for a single e-commerce store by its domain. Returns detailed info including traffic stats, product count, platform, social links, business model flags (dropshipping, print-on-demand), and more.
Inputs: domain (required).
Get advertisements detected for a specific store. Returns ad creatives with platform, image URL, landing URL, and date range when the ad was active.
Inputs: domain (required).
List products sold by a specific e-commerce store. Returns a paginated list of product titles, prices, images, and vendors.
Inputs: domain (required), page, sort, per_page.
Get the technology stack used by a store. Returns a list of technologies with their categories (e.g. analytics, payments, marketing, hosting).
Inputs: domain (required).
Get detailed traffic analytics for a store, including monthly visitors, trend percentage, bounce rate, average visit length, pages per visit, geographic breakdown, and traffic sources (direct, search, social, mail, display, referrals).
Inputs: domain (required).
Get trending products and rapidly growing stores. Useful for identifying hot products, emerging niches, and fast-rising stores.
Inputs: page, category, per_page.
Search for e-commerce advertisements across platforms. Filter by keyword or store domain. Returns ad creatives with platform, images, landing URLs, and active date ranges.
Inputs: page, sort, keyword, per_page, +2 more.
Search the Cart product database across all tracked stores. Filter by keyword, price range, and currency. Returns product titles, prices, images, store domain, and vendor.
Inputs: page, sort, keyword, currency, +3 more.
Search the Cart database of e-commerce stores. Filter by keyword, platform (e.g. shopify, woocommerce), language, minimum traffic, and more. Returns a paginated list of store summaries with traffic metrics, product counts, and platform info.
Inputs: page, sort, keyword, language, +3 more.
Search for product suppliers and vendors. Filter by keyword, location, or type. Useful for finding sourcing and manufacturing partners.
Inputs: page, sort, type, keyword, +2 more.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 15 tools
All 15 tools have distinct, non-overlapping purposes. For example, search_stores vs get_store search vs detail; get_store_products vs search_products differentiate by store-specific vs cross-store; and get_store_traffic, tech, ads each address unique aspects. No confusion between tools.
Every tool uses a consistent verb_noun snake_case pattern (e.g., search_stores, get_store, get_store_products, compare_stores). Verbs are appropriately chosen ('search', 'get', 'compare'), and the naming is uniformly descriptive and predictable.
With 15 tools, the server is well-scoped for its domain of e-commerce store and product intelligence. Each tool covers a specific operation without unnecessary duplication, and the count feels neither sparse nor overwhelming.
The tool set provides comprehensive coverage for the intended use case: retrieving store profiles, detailed analytics, product data, ads, supplier info, trends, and niche overviews. No obvious gaps exist—all major data types and operations (search, detail, compare, trending) are represented.
Hosted MCP for e-commerce: live product catalog, stock, and pricing for AI agents.
MCP server for building and testing AI agents with multi-model experimentation and insights.
Multi-tenant MCP gateway for AI commerce. One connection, every store.
Multi-tenant MCP gateway for AI commerce. One connection, every store.
