drio
Open app

Tavily MCP Server

Source

An MCP server for Tavily's search API

Catalog onlyCatalog onlySTDIO

Overview

Tavily MCP Server is a Model Context Protocol server designed to provide AI-powered web search capabilities using Tavily's search API. It enables large language models (LLMs) to perform sophisticated web searches, retrieve direct answers to user queries, and access recent news articles with AI-extracted relevant content.

To use the Tavily MCP Server, you install it via pip, configure it with your Tavily API key, and then run the server. You can perform various types of searches by sending specific commands through the server.

  • Extensive web search capabilities with AI-powered content extraction. - Direct answer generation with supporting evidence. - Recent news search functionality with publication dates.
  1. Conducting comprehensive web searches for research purposes.
  2. Gathering and synthesizing information for report writing.
  3. Tracking recent news developments on specific topics.

Add to your AI client

Use these steps to connect Tavily MCP Server in Cursor, Claude, VS Code, and other MCP-compatible apps. The same JSON appears in the Use with menu above for one-click copy.

Cursor

Add this to your .cursor/mcp.json file in your project root, then restart Cursor.

.cursor/mcp.json

{
  "mcpServers": {
    "mcp-tavily-ramxx": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-mcp-tavily-ramxx"
      ]
    }
  }
}

Claude Desktop

Add this server entry to the mcpServers object in your Claude Desktop config, then restart the app.

~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows)

{
  "mcpServers": {
    "mcp-tavily-ramxx": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-mcp-tavily-ramxx"
      ]
    }
  }
}

Claude Code

Add this to your project's .mcp.json file. Claude Code will detect it automatically.

.mcp.json (project root)

{
  "mcpServers": {
    "mcp-tavily-ramxx": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-mcp-tavily-ramxx"
      ]
    }
  }
}

VS Code (Copilot)

Add this to your .vscode/mcp.json file. Requires the GitHub Copilot extension with MCP support enabled.

.vscode/mcp.json

{
  "servers": {
    "mcp-tavily-ramxx": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-mcp-tavily-ramxx"
      ]
    }
  }
}

Windsurf

Add this to your Windsurf MCP config file, then restart Windsurf.

~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "mcp-tavily-ramxx": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-mcp-tavily-ramxx"
      ]
    }
  }
}

Cline

Open Cline settings, navigate to MCP Servers, and add this server configuration.

Cline MCP Settings (via UI)

{
  "mcpServers": {
    "mcp-tavily-ramxx": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-mcp-tavily-ramxx"
      ]
    }
  }
}

FAQ

**What programming language is used?**

Tavily MCP Server is built for Python, ensuring compatibility with various Python-based applications.

**Is an API key required?**

Yes, a Tavily API key is necessary for the server to access the search functionality. You can obtain one from Tavily's website.

**Can multiple searches be performed simultaneously?**

Yes, the server supports multiple requests, allowing for efficient searching across different queries.

Research and data