# Tavily MCP Server [Health: Active]

**Category:** 🔎 Search & Data Extraction  
**Repository:** https://github.com/Seey215/tavily-mcp  
**GitHub Stars:** 0  
**npm Downloads (last month):** 84909  
**Views:** 7  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/tavily-mcp-server

## Description
MCP server for advanced web search using Tavily API.

## Tools
Capabilities this server exposes over MCP:

- **tavily_search** — Search the web for current information on any topic. Use for news, facts, or data beyond your knowledge cutoff. Returns snippets and source URLs.
- **tavily_extract** — Extract content from URLs. Returns raw page content in markdown or text format.
- **tavily_crawl** — Crawl a website starting from a URL. Extracts content from pages with configurable depth and breadth.
- **tavily_map** — Map a website's structure. Returns a list of URLs found starting from the base URL.
- **tavily_research** — Perform comprehensive research on a given topic or question. Use this tool when you need to gather information from multiple sources to answer a question or complete a task. Returns a detailed response based on the research findings. Rate limit: 20 requests per minute.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "tavily-mcp-server": {
    "command": "npx",
    "args": ["-y","tavily-mcp@latest"],
    "env": {
      "TAVILY_API_KEY": ""
    }
  }
}
```

**Requires environment variables:** `TAVILY_API_KEY` — the values above are empty placeholders; fill in real credentials before running (see the repository for what each one is for).

## Documentation

## What Tavily MCP Server MCP server does

Tavily MCP Server MCP server gives an MCP-compatible AI client access to Tavily's web capabilities. Its tool set covers four related tasks: searching the web, extracting information from web pages, building a structured map of a website, and crawling a site to explore its pages.

The search capability is intended for current web information and supports filtering and domain-specific search features. Extraction focuses on retrieving useful content from supplied web pages. Mapping and crawling are suited to understanding a site's structure or collecting information across multiple pages.

The repository describes use with AI assistants, including Claude, and provides configuration guidance for Claude Desktop, Cursor, Cline, and VS Code.

## How it works

The server runs as an MCP process and uses the Tavily API for its web operations. An MCP client starts the process, passes the Tavily credential through the environment, and makes the exposed tools available to the assistant. The assistant can then call tools such as `tavily-search` and `tavily-extract` during a conversation.

The available capabilities are grouped around search and site exploration rather than a local content index. Tavily handles the web-facing operations, while the MCP layer makes those operations callable by the connected client.

## Setup and configuration

A Tavily API key is required. Create or obtain one through Tavily, then provide it as `TAVILY_API_KEY`. Node.js version 20 or newer is listed as a prerequisite. Git is only needed when using the Git-based installation method.

The simplest launch command is:

```bash
npx -y tavily-mcp@latest
```

Client configuration uses that command and passes the API key in the server environment. The README includes setup paths for Claude Desktop, Cursor, Cline, and VS Code. VS Code can use a user settings file or a workspace `.vscode/mcp.json` file, with the key entered through a password-protected input. Cline also supports installation from its marketplace. Cursor requires version 0.45.6 or newer and may require refreshing the server entry before its tools appear.

Tavily MCP Server MCP server can also be installed for Claude Desktop through Smithery using the Smithery CLI. A client is expected for practical use; launching the process by itself does not provide an interface for calling its tools.

## Tools and capabilities

- `tavily-search` for real-time web searches.
- `tavily-extract` for extracting information from web pages.
- A web mapping tool for creating a structured representation of a website.
- A web crawler for systematically exploring websites.
- Search filtering and domain-specific query options.

## Limitations and notes

The server depends on a Tavily API key and access to Tavily's services. It also requires Node.js 20 or later according to the documented prerequisites.

Client behavior varies by product. Cursor has the documented minimum version of 0.45.6, and the tool list may need a manual refresh after configuration. The examples use the latest npm package for NPX installation, while one VS Code installation link references a specific package version.

_Full upstream README: https://allmcps.com/mcp/tavily-mcp-server/readme_