Overview
Features
Breakdown queries across orders, carts and returns
Aggregations that summarize data as metrics (or statistics) on a field, with the result returned aggregated by a second field.
Date breakdown queries
Time-based aggregations that show the frequency of a date value over a time interval, applying an operator to a selected field.
Stats queries
Numeric computations performed on a specific field of a resource, returning the computed value.
Search queries
Retrieves actual matching records with sorting, filtering and pagination; not an aggregation.
Multi-resource support
Queries operate across the three resources—orders, carts and returns—using the four query types.
Ready-made use-case toolkit
Library of common metrics scenarios such as best-selling products by market, customers who bought a specific product, and frequently bought together items.
Local deployment and packaging
Build from source with pnpm, generate the dist folder, and deploy a local MCP server for use with AI agents.
Claude Desktop extension packaging
Package as a Claude Desktop extension (mcpb) and install by uploading the bundle, then connect using client_id and client_secret.
Who Is This For?
- Data analysts:Analyze orders, carts and returns metrics using the Metrics API via the MCP server.
- Developers integrating with AI agents:Connect the local MCP server to Claude Desktop or other AI agents to query metrics data.
- Business stakeholders / product teams:Leverage ready-made use cases to gain insights and support decision making.




