baidu-search-mcp

Baidu Search MCP Server

This is a Baidu search server implementation for Model Context Protocol (MCP), allowing AI assistants to perform intelligent searches using the Baidu Wenxin API.

Features

• Supports intelligent search using the Baidu Wenxin API
• Supports multiple model selections (ernie-3.5-8k, ernie-4.0-8k, deepseek-r1, deepseek-v3)
• Provides search results and reference sources
• Supports deep search and time-sensitive filtering

Installation

npm install @modelcontextprotocol/sdk axios

Configuration

1. First, you need to obtain the Baidu Wenxin API key:
   • Visit Baidu AI Cloud
   • Create an application and obtain the API key
2. Set environment variables:

export BAIDU_API_KEY=your_api_key_here

Usage

Run as a standalone server

node build/index.js

Use in MCP configuration

Add the following configuration to your MCP settings file:

{
  "mcpServers": {
    "baidu-search": {
      "command": "node",
      "args": ["/path/to/baidu-search-mcp/build/index.js"],
      "env": {
        "BAIDU_API_KEY": "your_api_key_here"
      },
      "disabled": false,
      "autoApprove": []
    }
  }
}

API

baidu_search

The search tool supports the following parameters:

• query (required): Search query text
• model: Name of the model to use
  • Optional values: "ernie-3.5-8k", "ernie-4.0-8k", "deepseek-r1", "deepseek-v3"
  • Default value: "ernie-3.5-8k"
• search_mode: Search mode
  • Optional values: "auto", "required", "disabled"
  • Default value: "auto"
• enable_deep_search: Whether to enable deep search (default: false)
• search_recency_filter: Time-sensitive range of search results
  • Optional values: "week", "month", "semiyear", "year"

Development

1. Clone the repository
2. Install dependencies:

npm install

3. Compile TypeScript:

npm run build

License

MIT License

Contributing

Issues and Pull Requests are welcome!

Disclaimer

This project is not responsible for the use of API keys. Please ensure compliance with the Baidu Wenxin API terms of use and policies.