GrokSearch

Level ALocal

Integrate Grok's powerful real-time search capabilities into Claude via the MCP protocol!

Overview

GrokSearch is a FastMCP-based MCP server with a dual-engine design: Grok handles AI-driven search while Tavily handles high-fidelity page fetching and site mapping, falling back to Firecrawl when Tavily fails. It gives Claude Code and other LLM clients real-time web access through tools like web_search, get_sources, web_fetch, and web_map, plus automatic time-context injection, model switching, and a one-click toggle to disable Claude Code's built-in search tools.

⭐ 1.8kSearch & InfoPythonRepo ↗Updated 2026-03-09

Installation

MCP client config (paste into mcp.json)

{
  "mcpServers": {
    "gudastudio-groksearch": {
      "command": "uvx",
      "args": [
        "--from",
        "git+https://github.com/GuDaStudio/GrokSearch@grok-with-tavily",
        "grok-search"
      ],
      "env": {
        "GROK_API_URL": "https://your-api-endpoint.com/v1",
        "GROK_API_KEY": "your-grok-api-key",
        "TAVILY_API_KEY": "tvly-your-tavily-key"
      }
    }
  }
}

Agent install prompt

Copy and paste into Cursor / Claude / other agents to install this MCP for you

Please read this MCP repository and install it into my current environment:
https://github.com/GuDaStudio/GrokSearch

Known working client config (can write into mcp.json):
{
  "mcpServers": {
    "gudastudio-groksearch": {
      "command": "uvx",
      "args": [
        "--from",
        "git+https://github.com/GuDaStudio/GrokSearch@grok-with-tavily",
        "grok-search"
      ],
      "env": {
        "GROK_API_URL": "https://your-api-endpoint.com/v1",
        "GROK_API_KEY": "your-grok-api-key",
        "TAVILY_API_KEY": "tvly-your-tavily-key"
      }
    }
  }
}

Please: 1) check README for deps and env vars; 2) write config for my client; 3) if secrets are missing, tell me what to fill — do not invent values.

GitHub

GrokSearch

Installation

Agent install prompt

Related servers