Web Search

A Model Context Protocol (MCP) server that provides a tool for performing Google searches and retrieving the content of the top results.

python

4

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "hexdecimal16-web-search-mcp": {
      "command": "uv",
      "args": [
        "--directory",
        "/path/to/your/project/web-search-mcp",
        "run",
        "web_search.py",
        "--timeout=15"
      ],
      "env": {
        "UV_DEBUG": "false"
      }
    }
  }
}

You can run the Web Search MCP Tool as an MCP server to perform Google searches, fetch the content of the top results, and return the aggregated content to your MCP client. This is useful for quickly integrating live web content into your workflows or assistants through a standardized MCP interface.

How to use

Launch the Web Search MCP Tool from your MCP client using a stdio transport. Once running, you provide a search query to the tool, which then performs a Google search, identifies the top 5 non-social results, crawls their content, and returns all crawled content as a single string. You can use this to feed web content into your prompts, build dynamic knowledge, or verify information from current web sources.

How to install

Prerequisites: You need Python 3.13 or newer to run the tool and uv for package management.

Install uv if it is not already installed.

Then clone the project repository and set up a Python environment, install dependencies, and prepare to run.

Run the following commands exactly as shown to install and prepare the environment.

Configuration and starting the tool

{
  "mcpServers": {
    "web_search": {
      "command": "uv",
      "args": [
        "--directory",
        "/path/to/your/project/web-search-mcp",
        "run",
        "web_search.py",
        "--timeout=15"
      ]
    }
  }
}

Testing and validation

You can verify the script is syntactically correct and the tool is functioning by running the test suite with pytest after installing test dependencies.

uv pip install pytest pytest-asyncio
pytest

Available tools

google_search

Performs a Google search for a given query and returns search results.

identify_top_results

Identifies the top 5 non-social results from the search results.

crawl_results

Crawls the content of the top results to gather textual content.

return_crawled_content

Returns the crawled content as a single concatenated string for easy consumption by the MCP client.