mcp-amazon

This MCP server is designed to help Amazon sellers evaluate potential products by estimating costs, revenues, and profit margins. It analyzes input product data (such as cost of goods, Amazon fees, shipping costs, and estimated sales price) to generate profitability metrics, helping you compare candidates quickly. The server exposes a set of MCP tools that let you calculate unit economics, run scenario analyses (e.g., different price points or cost assumptions), and export summarized results for decision making. To use it, provide the required product inputs to the MCP interface, choose the analysis you want (cost breakdown, profit forecast, break-even analysis), and review the generated insights to identify promising products.

Key capabilities typically include cost decomposition (COGS, FBA/fulfillment fees, shipping, duties), price-to-profit calculations, ROI and net profit projections, and scenario planning (varying cost or price inputs). The tools are designed to be scriptable via the MCP protocol, so you can integrate them into your product research workflows, dashboards, or batch analyses.

If you are using the recommended packaging (via npx), simply start the server in your project directory and interact with its endpoints or commands as described by the MCP interface. The server is intended to streamline supplier cost data, listing fees, and revenue assumptions into a coherent profitability picture.

Prerequisites:

• Node.js v14+ and npm installed on your machine
• Internet access to fetch the MCP package

Option A: Run with npx (no local installation)

1. Ensure Node.js and npm are installed
2. Run the MCP package directly without installing globally:

npx -y airskywater-mcp-amazon-server

Option B: Install as a local package (optional but recommended for offline usage)

1. Initialize a project (if needed) and install the package:

mkdir mcp-project && cd mcp-project
npm init -y
npm install airskywater-mcp-amazon-server

2. Start or invoke the server through your project scripts according to the MCP interface provided by the package.

Prerequisites recap:

• Node.js installed
• Access to npm (or npx)
• (Optional) Internet access for initial installation

Note: If the package provides environment variable configuration, you can prepare an .env file or export variables in your shell before starting the server.

Environment variables and configuration:

• You may need to provide API keys or credentials for data sources (e.g., supplier cost data, pricing feeds). Add these as env vars per the package documentation, for example: export AMAZON_API_KEY=your-key or set in .env.
• Common options include currency, tax/VAT handling, and default cost assumptions. If the MCP package supports profiles, create a profile with your typical margins and fees to speed up analyses.
• If you encounter rate limits or slow responses, consider running analyses in batch with time gaps or use a cached data approach if supported.
• When using npx, ensure you’re connected to the internet for the initial fetch; after installation, ensure the package version aligns with your project’s needs.
• For troubleshooting, check the MCP server logs for missing data inputs or validation errors and ensure all required fields (costs, fees, and revenue) are provided in the analysis requests.