home / mcp / websearch mcp server

WebSearch MCP Server

Web Search tools are a series of tools that allow Claude to acces de internet via MCP Server

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "josemartinrodriguezmortaloni-websearch-tools": {
      "command": "uv",
      "args": [
        "--directory",
        "D:\\ABSOLUTE\\PATH\\TO\\WebSearch",
        "run",
        "main.py"
      ],
      "env": {
        "OPENAI_API_KEY": "YOUR_OPENAI_API_KEY_HERE",
        "FIRECRAWL_API_KEY": "YOUR_FIRECRAWL_API_KEY_HERE"
      }
    }
  }
}

WebSearch MCP Server provides advanced web search, content extraction, web crawling, and scraping capabilities exposed as an MCP server you can run locally or integrate with compatible clients. It enables automated information gathering and analysis through a simple, scriptable interface that can be driven by MCP Clients to perform targeted searches, extract details from pages, crawl sites, and scrape content for downstream use.

How to use

You access WebSearch via an MCP client by starting the local MCP runner and pointing your client to the WebSearch server configuration. Once running, you can issue the available endpoints as actions to perform searches, extract specific information from pages, crawl sites to collect content, or scrape pages for structured data.

How to install

{
  "mcpServers": {
    "websearch": {
      "type": "stdio",
      "name": "websearch",
      "command": "uv",
      "args": [
        "--directory",
        "D:\\ABSOLUTE\\PATH\\TO\\WebSearch",
        "run",
        "main.py"
      ],
      "env": {
        "FIRECRAWL_API_KEY": "YOUR_FIRECRAWL_API_KEY_HERE",
        "OPENAI_API_KEY": "YOUR_OPENAI_API_KEY_HERE"
      }
    }
  }
}

Additional notes

Prerequisites include Python 3.8 or higher and the uv package manager. You will also need API keys for the Firecrawl service and OpenAI if you plan to use AI-powered features.

You can set up environment variables in a .env file at the project root to ensure keys are loaded automatically when the server starts.

Configuration and environment

The server start configuration uses uv to run the main Python script from a specified directory. The example shows how to point to your WebSearch project folder and run the entry script.

Troubleshooting

If you encounter issues, verify that the required API keys are set and loaded, confirm the working directory path is correct, and ensure the uv runner is accessible in your PATH. Check that the environment variables are available to the process starting WebSearch.

Available tools

Search

Perform intelligent web searches with customizable parameters and receive results in JSON.

Extract

Extract specific information from web pages using prompts and optional web search enhancement.

Crawl

Crawl websites with configurable depth and page limits, returning content in Markdown/HTML.

Scrape

Scrape content from target pages, with optional screenshots and formats.