Web Search MCP
A Model Context Protocol (MCP) server that provides web search functionality by scraping Google, DuckDuckGo, and Bing search results through a headless browser.
接入你的智能体
从发布者处安装。不是 PlugMCP 端点。
远程 URL 还是本机命令? 两种安装方式的区别
来自发布者
A Model Context Protocol (MCP) server that provides web search functionality by scraping Google, DuckDuckGo, and Bing search results through a headless browser.
A Model Context Protocol (MCP) server that provides web search functionality using a headless Chrome browser to scrape Google, DuckDuckGo and Bing search results.
search_webSearch the web across multiple engines (Google → DuckDuckGo → Bing) with automatic fallback and return structured results.
Parameters:
query (string): The search query stringmax_results (int, optional): Maximum number of results to return (default: 10, max: 100)include_snippets (bool, optional): Whether to include text snippets (default: true)Returns:
title: Page titleurl: Full URL (redirect wrappers resolved to the real destination)domain: Domain namesnippet: Text snippet (if enabled)rank: Search result rankingsource_engine: Which engine produced the result (google, duckduckgo, or bing)get_webpage_contentFetch and return the text content of a webpage.
Parameters:
url (string): The URL of the webpage to fetchmax_length (int, optional): Maximum content length (default: 5000, max: 20000)Returns:
url: The requested URLtitle: Page titlecontent: Extracted text contentlength: Content length in charactersget_search_engine_statusReport the active browser backend and the availability of each search engine.
Returns:
backend key (selenium or obscura) plus each engine name mapped to available or blocked.reset_search_enginesClear the list of engines marked as blocked during the session (useful if an engine recovers).
Returns:
Install dependencies:
# Using uv (recommended)
uv sync
# Or using pip
pip install -e .
Install a browser backend (pick one):
Selenium (default) — requires Chrome:
brew install --cask google-chromesudo apt-get install google-chrome-stableChromeDriver is downloaded and managed automatically by webdriver-manager.
Obscura — no Chrome required. See the Browser Backends section below.
# Run directly
python main.py
# Or using the installed script
web-search-mcp
The server will start and listen for MCP connections.
Add this configuration to your Claude Desktop MCP settings:
{
"mcpServers": {
"web-search-mcp": {
"command": "uvx",
"args": ["git+https://github.com/pranavms13/web-search-mcp"]
}
}
}
Once connected, you can use the tools like this:
Search for "python web scraping tutorials" and show me the top 5 results.
Get the content from this webpage: https://example.com/article
The web searcher uses these Chrome options by default:
The server can render pages with one of two interchangeable backends, selected via the BROWSER_BACKEND environment variable:
Both backends produce fully JS-rendered HTML, which is then parsed with BeautifulSoup, so search results are identical in shape regardless of backend.
Install the Obscura binary (see its releases), e.g. on macOS:
curl -LO https://github.com/h4ckf0r0day/obscura/releases/latest/download/obscura-x86_64-macos.tar.gz
tar xzf obscura-x86_64-macos.tar.gz
Run the server with Obscura selected:
BROWSER_BACKEND=obscura web-search-mcp
If obscura is not on your PATH, point to it explicitly with OBSCURA_BIN=/full/path/to/obscura. When the binary cannot be found, the server logs a warning and falls back to the Selenium backend automatically.
The active backend is reported by the get_search_engine_status tool under the backend key.
The tool includes comprehensive error handling for:
Errors are logged and graceful fallbacks are provided.
fastmcp: MCP server frameworkselenium: Web browser automation (Selenium backend)beautifulsoup4: HTML parsingwebdriver-manager: Chrome driver management (Selenium backend)lxml: XML/HTML parserThe Obscura backend has no Python dependency — it is invoked as an external CLI binary.
To modify or extend the functionality:
uv sync or pip install -e .uv run pytest test.py -qpython main.py (or BROWSER_BACKEND=obscura OBSCURA_BIN=/path/to/obscura python main.py)This project is licensed under MIT License. You can check it out at - LICENSE
Contributions are welcome! Please feel free to submit a Pull Request.
A Model Context Protocol (MCP) server that provides web search functionality using a headless Chrome browser to scrape Google, DuckDuckGo and Bing search results.
工具
暂无工具列表。远程服务器在首次健康检查后会获得工具列表。
常见问题
在本页复制适用于 Cursor、Claude Desktop、Claude Code 或 VS Code 的 stdio 配置。它使用发布者的安装方式,而不是 PlugMCP 代理。
此记录未列出必需的环境变量。
Web Search 通过 stdio 在本地运行。
上次检查时间:2026-10-28。 A Model Context Protocol (MCP) server that provides web search functionality using a headless Chrome browser to scrape Google, DuckDuckGo and Bing search results.