Amazon Data MCP

Amazon AI MCP is a data access layer built on the Model Context Protocol that allows AI agents to query live Amazon marketplace information programmatically. It operates over a streamable-HTTP endpoint, meaning compatible AI applications can retrieve structured product data without building custom scrapers or maintaining separate API integrations.

The server provides access to product listings, customer reviews, Best Sellers Rank data, seller profiles, category hierarchies, and keyword search trends. Responses are formatted as JSON objects that AI agents can parse and reason over directly, removing the need for intermediate data transformation.

This approach differs from traditional Amazon data APIs in that it targets agent consumption rather than human-facing dashboards. Developers using MCP-compatible tools such as Claude, Cursor, or Claude Code can configure the endpoint once and access all available data queries through the protocol's standard tool-calling mechanism.

The tool is maintained as an open-source project on GitHub under the Pangolin-spg organization and is distributed as a Python package on PyPI. Access requires a permanent API key, with a free tier available for testing.