Bocha Ai MCP Server

🚀 Bocha AI Web Search MCP Server

A Model Context Protocol (MCP) server dedicated to the web search functionality of Bocha AI. This server offers a powerful search tool that retrieves web page content and relevant images, supporting both human-readable Markdown output and raw JSON data formats.

🚀 Quick Start

The Bocha AI Web Search MCP Server is a specialized server designed to serve the web search function of Bocha AI. It provides a robust search tool that can retrieve web page content and relevant images from the Chinese Internet. The server supports both Markdown, a human - readable format, and JSON, a structured data format for programmatic processing.

✨ Features

• Web Search: Retrieve relevant web page content, summaries, and related images from the Chinese Internet.
• Flexible Time Range: Support search results within a time frame from the past day to an unlimited period.
• Intelligent Summaries: Optionally generate detailed text summaries.
• Dual - Format Output:
  • Markdown: A human - readable format with links and image previews.
  • JSON: Structured data suitable for programmatic processing.
• Robust Error Handling: Comprehensive error management with clear message prompts.

📦 Installation

Make sure you have Node.js and npm installed.

git clone git@github.com:yoko19191/bocha-ai-mcp-server.git
cd bocha-ai-mcp
cp .env.example .env # Configure your Bocha AI API key
npm install
npm run build

📚 Documentation

About Bocha AI

Bocha AI (Bocha AI) is a legal and compliant artificial intelligence search engine operating in China. It focuses on providing search services for Chinese Internet content, offering comprehensive search capabilities while strictly adhering to relevant Chinese laws and standards.

Tools

bocha_web_search

A powerful web search tool that retrieves content from Chinese Internet sources.

• query (Required):
  • Type: String
  • Description: The search term to look for.
• freshness (Optional):
  • Type: Enum
  • Values: "OneDay", "OneWeek", "OneMonth", "OneYear", "noLimit"
  • Default: "noLimit"
  • Description: Filter results by time.
• summary (Optional):
  • Type: Boolean
  • Default: false
  • Description: Whether to generate a comprehensive summary for search results.
• count (Optional):
  • Type: Number
  • Range: 1 - 50
  • Default: 10
  • Description: Control the number of returned results.
• raw_json (Optional):
  • Type: Boolean
  • Default: false
  • Description: When false, return human - readable Markdown; when true, return structured JSON data.

Configuration

mcp.json

{
  "mcpServers": {
    "bocha-ai-search": {
      "command": "node",
      "args": [
        "/ABSOLUTE/PATH/bocha-ai-mcp-server/build/index.js"
      ],
      "env": {
        "BOCHA_API_KEY": "<YOUR_BOCHA_API_KEY>"
      }
    }
  }
}

cursor

env BOCHA_API_KEY=YOUR_BOCHA_API_KEY node /ABSOLUTE/PATH/bocha-ai-mcp-server/build/index.js

Future Work Plan

The project currently implements the web search API functionality of Bocha AI. Future updates will include:

• AI Search: Advanced semantic search capabilities.
• Proxy Search: Intelligent proxy search functionality.

📄 License

MIT