Mcp4amazon

🚀 Amazon Data Analysis Project

This project is designed for Amazon operators, enabling large models to directly read and analyze data from Amazon.

🚀 Quick Start

This project is tailored for Amazon operations, allowing large models to directly read and analyze data from Amazon. Here are the current features and how to set up and use the project.

✨ Features

get_product_info

Retrieve product information for a given ASIN, including the product title, description, and all variant details.

• Parameters:
  • asin: Amazon ASIN code

get_product_review

Fetch review information for a given ASIN, with the total number of characters not exceeding max_length.

• Parameters:
  • asin: Amazon ASIN code
  • max_length: Total number of characters

📦 Installation

Python and git clone

The installation of Python and the git clone operation are not detailed here. The Python version used in this project is 3.11.

Install uv

Open the CMD and run the following command:

pip install uv

Create a virtual environment

Navigate to your project directory and create a virtual environment:

cd E:\MyProject\MCP4Amazon (Replace with your actual project path)
uv venv
.\.venv\Scripts\activate

Install dependencies

Install the project dependencies using the following command:

uv pip install -r requirements.txt

🔧 Configuration

Configure Google Chrome

This project uses Playwright to operate Google Chrome through port 9222 to obtain data.

1. Right-click on the Google Chrome shortcut on the desktop and select "Properties".
2. Add the following characters to the "Target" field:

--remote-debugging-port=9222

For example:

"D:\Program Files\chrome-win64\chrome.exe" --remote-debugging-port=9222

3. Save the changes and double-click the shortcut to open the browser.

Use Cherry Studio to call the model

Among the free models, only Claude Desktop has an MCP interface, but it is not available in China. Currently, the feasible solutions are to use Cherry Studio to call the large model API (requires payment) or use Ollama.

Configure the MCP server

{
    "mcpServers": {
        "MCP4Amazon": {
            "type": "stdio",
            "command": "uv",
            "args": [
                "--directory",
                "E:\\MyProject\\MCP4Amazon",
                "run",
                "mcp4amazon.py"
            ]
        }
    }
}

1. Add a server -> Import from JSON -> Copy and paste the above text.
2. Replace the second line in args with your actual project path.
3. Save the configuration.

Configure the large model API or Ollama

Refer to the Cherry Studio tutorial for specific operations.

📈 Results