MCP Server Product Name: Bocha

Version Information

v1

Product Description

Short Description

Bocha is a search engine for AI, enabling your AI applications to access high-quality world knowledge from nearly 10 billion web pages and content sources, covering various fields such as weather, news, encyclopedia, healthcare, train tickets, and images.

Long Description

Bocha is a search engine for AI, enabling your AI applications to access high-quality world knowledge from nearly 10 billion web pages and content sources, covering various fields such as weather, news, encyclopedia, healthcare, train tickets, and images.

Category

Web Search

Tags

Search, News, Weather, Encyclopedia

Tools

Tool1: Bocha Web Search

Detailed Description

Search the entire web and web links via Bocha, returning results that include web page titles, URLs, summaries, website names, website icons, publication dates, and image links.

Parameters Required for Debugging

Input:

• query: Search term (required)
• freshness: Search within a specified time range (optional values YYYY-MM-DD, YYYY-MM-DD..YYYY-MM-DD, noLimit, oneYear, oneMonth, oneWeek, oneDay. Default is noLimit)
• count: Number of results to return (1-50, default is 10)

Output:

• Web page title, URL, summary, publication date, website name

Tool2: Bocha AI Search

Detailed Description

Building on Bocha's web search, AI identifies the semantics of the search terms and additionally returns structured modal cards for specific verticals, such as weather cards, calendar cards, and encyclopedia cards, among dozens of others. This provides better performance in semantic recognition, timeliness of search results, and richness of content.

Parameters Required for Debugging

Input:

• query: Search term (required)
• freshness: Search within a specified time range (optional values YYYY-MM-DD, YYYY-MM-DD..YYYY-MM-DD, noLimit, oneYear, oneMonth, oneWeek, oneDay. Default is noLimit)
• count: Number of results to return (1-50, default is 10)

Output:

• Web page title, URL, summary, publication date, website name, modal card

Compatible Platforms

Ark, Python, Claude, Cursor, etc.

Service Activation Link

You need to visit the Bocha AI Open Platform, log in, and obtain an API KEY.

Authentication Method

API Key

Installation and Deployment

Step 1: Download the Code to Your Local Machine

bash
git clone git@github.com:BochaAI/bocha-search-mcp.git

Step 2: Configure in the Client

Claude Desktop

On MacOS: ~/Library/Application Support/Claude/claude_desktop_config.json

On Windows: %APPDATA%/Claude/claude_desktop_config.json

json
"mcpServers": {
"bocha-search-mcp": {
"command": "uv",
"args": [
"--directory",
"/path/to/bocha-search-mcp",
"run",
"bocha-search-mcp"
],
"env": {
"BOCHA_API_KEY": "sk-****"
}
}
}

Step 3: Use in the Client

Step 4: Debug the Local Service (Optional)

bash
npx @modelcontextprotocol/inspector uv --directory /path/to/bocha-search-mcp run bocha-search-mcp

Customer Cases

Bocha has served over 3,000 enterprise users and 20,000 developer users, becoming the official networked search provider for DeepSeek and the recommended search API by Alibaba, Tencent, and ByteDance. Currently, it handles over 60% of the networked search requests for AI applications in China.

The content sources for Bocha's search include nearly 10 billion web pages and ecosystem partner content (including short videos, news, encyclopedias, weather, healthcare, train tickets, hotels, restaurants, attractions, businesses, and academic resources). Bocha will continue to collaborate with various platforms on content ecosystems and intelligent agent creation, providing rich and diverse answers to search queries for its users.

Frequently Asked Questions

What capabilities does the Bocha Web Search API service provide?Bocha Web Search provides a universal web search capability. You can use Bocha to search for information and web page links across the entire web, with the returned results including web page titles, web URLs, web page summaries, website names, website icons, publication times, image links, etc. Each search result can return up to 50 web pages (count50).

Traditional search engines use a keyword + bid ranking mechanism in their search algorithms. The goal of the search results is not to directly provide the correct answers to users but to attract user clicks to generate advertising revenue.

Bocha is a next-generation search engine based on multimodal hybrid search and semantic ranking technology. It supports natural language search methods for AI application scenarios, and the objective of the search results is to provide clean, accurate, and high-quality answers.

Bocha's semantic ranking technology is based on the Transformer architecture, which ranks search results according to their semantic relevance to the user's query. Since large models also use the Transformer architecture and make decisions based on the semantic relevance between context and user queries, these large models tend to prefer the search results provided by Bocha.

Currently, Bocha's search performance is the closest to that of the Bing Search API among domestic search engines. Because the Bing Search API data is hosted overseas (no domestic region), is expensive (15 USD per thousand requests), and does not provide text summaries (only 50-100 character snippets), many domestic enterprise customers have switched from Bing to Bocha.

What capabilities can the Bocha AI Search API service provide?

Bocha AI Search, built on top of Bocha Web Search, uses AI to recognize the semantics of search terms and returns additional structured modal cards for vertical content, such as weather cards, calendar cards, encyclopedia cards, and dozens of other types of modal cards. This improves the semantic recognition, timeliness of search results, and richness of content.

The currently supported types of modal cards include: weather, encyclopedia, medical, perpetual calendar, train, zodiac, precious metals, exchange rates, oil prices, mobile phones, stocks, and cars, among others.

For example, in the case of stock information, real-time stock data is generally not available on web pages and requires structured modal cards for support. The Bocha AI Search API can, in addition to providing web page information, output structured data cards for stock prices. Through the structured data provided by these modal cards, the accuracy of AI applications in answering time-sensitive questions can be further enhanced.