Baidu Search MCP Server

@Evilran

About Baidu Search MCP Server

Baidu Search MCP Server I A Model Context Protocol (MCP) server that provides web search capabilities through Baidu, with additional features for content fetching and parsing.

#baidu #mcp #search #web-search #content-fetching

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "baidu-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "@smithery/cli",
        "install",
        "@Evilran/baidu-mcp-server",
        "--client",
        "claude"
      ]
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is Baidu Search MCP Server?

A Model Context Protocol (MCP) server that provides web search capabilities through Baidu, with additional features for content fetching and parsing. It is designed for developers and AI applications that need LLM-friendly search results from Baidu.

How to use Baidu Search MCP Server?

Install via pip (uv pip install baidu-mcp-server) or Smithery, then configure it as an MCP server in Claude Desktop by adding a baidu-search entry with the command uvx baidu-mcp-server in the MCP configuration file. The server exposes two tools: search (for web searches) and fetch_content (for webpage content extraction).

Key features of Baidu Search MCP Server

Web search on Baidu with configurable result count (up to 10)
Intelligent content fetching and text extraction from webpages
Built-in rate limiting: 30 requests/min for search, 20 for fetching
Ad removal, redirect URL cleaning, and LLM-friendly output formatting
Comprehensive error handling and logging via MCP context
Easy installation via PyPI, uv, or Smithery

Use cases of Baidu Search MCP Server

Enabling LLMs like Claude to search Baidu for real-time information
Retrieving and processing webpage content for AI-assisted research
Building AI agents that need access to Baidu search results without ads or redirects
Automating content extraction for data analysis or summarization tasks

FAQ from Baidu Search MCP Server

What dependencies or runtime are required?

The server is a Python package that requires Python and the uv package manager (or pip). It runs as a command-line MCP server, typically launched by a client like Claude Desktop.

Where does search data come from and where is it stored?

All search results and fetched content are retrieved live from Baidu. The server does not cache or store any data persistently.

What are the known limits?

Search requests are limited to 30 per minute, and content fetching to 20 per minute. The server automatically queues requests and applies wait times to respect rate limits.

How is the server accessed and what transport does it use?

The server is accessed via the MCP protocol using stdio transport (command-line invocation). No authentication is required to run the server; it connects directly to Baidu’s public web search.

How does this server differ from alternatives?

While many MCP servers use DuckDuckGo or other search engines, this server is specifically built for Baidu, making it ideal for users in China or those needing Baidu-specific results and redirect cleaning.

Frequently asked questions