MCP Server: E-Commerce Analytics

An MCP (Model Context Protocol) server that gives Claude direct access to an e-commerce SQLite database. Ask natural-language questions about sales, customers, inventory, and orders -- Claude calls the right tools automatically.

Built with the official Anthropic MCP Python SDK and designed as a single-file, zero-config demo. The database is created and seeded with realistic data on first run.

Features

| Tool | Description | |---|---| | query_sales_report | Revenue, order count, average order value, and top products for any date range. Optional category filter. | | get_customer_insights | Full customer profile: purchase history, lifetime spend, favourite categories, recent orders. Look up by ID or email. | | check_inventory_status | Stock levels for every product with low-stock alerts. Optional category filter and configurable threshold. | | search_orders | Flexible order search by status, customer email, and minimum amount. Paginated results with item counts. | | get_business_summary | High-level dashboard: total revenue, revenue by category, 6-month trends, and top 5 customers. |

Quick Start

1. Install dependencies

pip install -r requirements.txt

2. Run the server directly (stdio transport)

python server.py

On first run the server creates ecommerce_demo.db in the same directory and seeds it with:

50 products across 5 categories (Electronics, Clothing, Home & Kitchen, Books, Sports)
200 customers with realistic names, emails, and signup dates
1 200 orders with varied statuses and 1-5 line items each

3. Connect to Claude Desktop

Add the following to your Claude Desktop configuration file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "ecommerce-analytics": {
      "command": "python",
      "args": ["/absolute/path/to/server.py"]
    }
  }
}

Restart Claude Desktop. The tools will appear automatically.

Example Queries

Once connected, try asking Claude:

"Show me a sales report for the last 30 days."
"How is the Electronics category performing this quarter?"
"Look up the customer with email james.smith@example.com."
"Which products are running low on stock?"
"Find all pending orders over $100."
"Give me an overall business summary with trends."

Claude will select the appropriate tool, execute the query, and present the results in a readable format.

Tech Stack

Python 3.10+
MCP Python SDK (mcp package with FastMCP)
Pydantic v2 for response validation
SQLite (zero-config, file-based)

Project Structure

mcp-server-ecommerce-analytics/
  server.py           # MCP server -- tools, DB setup, entry point
  requirements.txt    # Python dependencies
  README.md           # This file
  ecommerce_demo.db   # Auto-generated on first run

License

MIT

MCP Servers