BigGo-MCP-Server 0.1.1

A Model Context Protocol (MCP) server that provides product search, price history tracking, and specification search capabilities.

🔍 BigGo MCP Server

A Model Context Protocol (MCP) server that provides product search, price history tracking, and specification search capabilities.

✨ Features

🛍️ Product Search

Search for products across multiple e-commerce platforms with natural language queries.

Example Prompts:

Find me iPhone 15 Pro on Shopee

Look for Nike running shoes

📈 Price History Tracking

Track product price history in two ways:

• Direct URL tracking
• History ID tracking (obtained from product search results)

Example Prompts:

Show me the price history of this product: https://www.momoshop.com.tw/goods/GoodsDetail.jsp?i_code=13660781

Find me the price history of IPhone 15 Pro at Shopee?

📐 Product Specification Search

Search for products based on specific technical specifications using Elasticsearch.

Example Prompts:

Find me refrigerators with the highest energy efficiency rating

Search for diving watches that weigh around 120g

Show me Japanese air conditioners with heating capability and low noise levels

🛠️ Available Tools

🛍️ Product Search Tools

• product_search
  • Product search with biggo search api

📈 Price History Tools

• price_history_graph
  • Link that visualizes product price history
• price_history_with_history_id
  • Uses history IDs from product search results
  • Typical tool workflow: Search product -> Get history ID -> Track prices
• price_history_with_url
  • Tracks price history using product URLs

📐 Specification Search Tools

• spec_indexes
  • Lists available Elasticsearch indexes for product specifications
• spec_mapping
  • Shows Elasticsearch index mapping with example documents
• spec_search
  • Advanced specification-based product search

🧰 Utility Tools

• get_current_region
  • Get the current region setting

⚙️ Installation

📋 Prerequisites

PyPi package link: biggo-mcp-server

1. Python 3.13 or higher
2. uv package manager
3. BigGo API credentials (client ID and secret) for specification search. Available at BigGo Account

💻 From Local Project

Use absolute path for --directory argument.

{
  "mcpServers": {
    "biggo-mcp-server": {
      "command": "uv",
      "args": [
        "run",
        "--directory",
        "/path/to/biggo-mcp-server",
        "biggo-mcp-server",
        "--client-id",
        "YOUR_CLIENT_ID",
        "--client-secret",
        "YOUR_CLIENT_SECRET"
      ],
      "enabled": true
    }
  }
}

📦 From Published Package

For specific version use biggo-mcp-server@VERSION, ex: biggo-mcp-server@0.0.1-dev3

{
  "mcpServers": {
    "biggo-mcp-server": {
      "command": "uvx",
      "args": [
        "biggo-mcp-server",
        "--client-id",
        "YOUR_CLIENT_ID",
        "--client-secret",
        "YOUR_CLIENT_SECRET"
      ],
      "enabled": true
    }
  }
}

🔧 Configuration Arguments

Variable	Description	Default	Choices
--region	Region for product search	TW	US, TW, JP, HK, SG, MY, IN, PH, TH, VN, ID
--client-id	Client ID	None	Required for specification search
--client-secret	Client Secret	None	Required for specification search
--log-level	Log level	INFO	DEBUG, INFO, WARNING, ERROR, CRITICAL
--es-proxy-url	Elasticsearch proxy URL	https://api.biggo.com/api/v1/mcp-es-proxy/	Any valid URL
--es-verify-certs	Verify Elasticsearch certificates	True	True, False
--auth-token-url	Auth token URL	https://api.biggo.com/auth/v1/token	Any valid URL
--auth-verify-certs	Verify Auth certificates	True	True, False

👨‍💻 Development

🚀 Setup

1. Install uv package manager
2. Install dependencies:

   uv sync
   

🧪 Testing and Development

1. Run with MCP Inspector:

   npx @modelcontextprotocol/inspector uv run biggo-mcp-server
   

2. Run tests:

   uv run --group test pytest
   

📦 Project Architecture

src/
└── biggo_mcp_server/
    ├── __init__.py         # MCP Server Entrypoint
    ├── lib/
    │   ...
    │   ├── server.py       # Server class      
    │   └── server_setup.py # Server initialization (load tools..etc)
    ├── services/           # Tool logic
    ├── tools/              # Tool entrypoint
    └── types/
        ├── api_ret/        # API responses
        ...
        ├── responses.py    # Tool responses
        └── setting.py      # Server setting

📄 License

This project is licensed under the MIT License. See the LICENSE file for details.