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.

Features

• Multi-Engine Search: Searches Google, DuckDuckGo, and Bing with automatic fallback when an engine is blocked or returns nothing
• Structured Results: Titles, URLs, domains, snippets, and rankings — with redirect URLs unwrapped to their real destinations
• Web Page Content: Fetch and extract text content from any webpage
• Pluggable Browser Backends: Selenium + headless Chrome (default) or the lightweight Obscura headless browser, selected via an environment variable
• MCP Compatible: Fully compatible with Claude Desktop and other MCP clients

Tools Available

search_web

Search the web across multiple engines (Google → DuckDuckGo → Bing) with automatic fallback and return structured results.

Parameters:

• query (string): The search query string
• max_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:

• List of search results with:
  • title: Page title
  • url: Full URL (redirect wrappers resolved to the real destination)
  • domain: Domain name
  • snippet: Text snippet (if enabled)
  • rank: Search result ranking
  • source_engine: Which engine produced the result (google, duckduckgo, or bing)

get_webpage_content

Fetch and return the text content of a webpage.

Parameters:

• url (string): The URL of the webpage to fetch
• max_length (int, optional): Maximum content length (default: 5000, max: 20000)

Returns:

• Dictionary with:
  • url: The requested URL
  • title: Page title
  • content: Extracted text content
  • length: Content length in characters

get_search_engine_status

Report the active browser backend and the availability of each search engine.

Returns:

• Dictionary with a backend key (selenium or obscura) plus each engine name mapped to available or blocked.

reset_search_engines

Clear the list of engines marked as blocked during the session (useful if an engine recovers).

Returns:

• A status message string.

Installation

1. Install dependencies:

   # Using uv (recommended)
   uv sync
   
   # Or using pip
   pip install -e .
   

2. Install a browser backend (pick one):

   • Selenium (default) — requires Chrome:
     • On macOS: brew install --cask google-chrome