Tavily Search MCP Server

@apappascs

Repository

Basic information

About Tavily Search MCP Server

An MCP server implementation that integrates the Tavily Search API, providing optimized search capabilities for LLMs.

#tavily #search #mcp #server #llm #api

Tavily Search MCP Server config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "tavily-search-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "@smithery/cli",
        "install",
        "@apappascs/tavily-search-mcp-server",
        "--client",
        "claude"
      ]
    }
  }
}

Tavily Search MCP Server tools

Execute web searches using the Tavily Search API.

Overview

What is Tavily Search MCP Server?

Tavily Search MCP Server is an MCP server implementation that integrates the Tavily Search API, providing optimized web search capabilities for large language models.

How to use Tavily Search MCP Server?

Clone the repository, install dependencies, build the project, and configure your Tavily API key in the Claude Desktop MCP configuration file. Use the tavily_search tool to execute web searches with optional parameters like search depth, topic, domain filtering, and result inclusion of images or raw content.

Key features of Tavily Search MCP Server

Web search optimized for LLMs
Content extraction from search results
Optional images, image descriptions, and LLM-generated answers
Domain filtering (include or exclude domains)
Control over search depth, topic, and time range

Use cases of Tavily Search MCP Server

Enhancing LLM responses with real-time web data
Aggregating news articles with time range filtering
Extracting relevant content from search results for analysis
Filtering searches by specific domains for focused research
Including images and descriptions for rich information retrieval

Frequently asked questions

Do I need a Tavily API key?

Yes, a Tavily API key is required. Obtain one from the Tavily dashboard after signing up.

What are the runtime dependencies?

Node.js is required for npm-based usage; Docker is also supported. The server can run via stdio or SSE transport.

Where does the search data come from?

All data is fetched from the Tavily Search API. No search results are stored locally.

What transports are supported?

The server supports both stdio (default) and SSE transport, configurable in the Claude Desktop configuration or via the `TRANSPORT` environment variable when using Docker.

Is there any limit on the number of results?

The `max_results` parameter defaults to 5 and can be increased, but actual limits depend on your Tavily API plan (a free tier is available).

Comments

More Search MCP servers

Knowra

Rex Zhang

Search papers on arXiv by research topic and exact announcement date, then read full-text paper briefs. An arXiv paper search and literature search tool for AI agents: search_papers returns titles, authors and abstracts

Default Privacy

Sourced privacy answers for AI assistants: how private an app or service is, private replacements for mainstream apps, privacy tool facts with sources, guides, MCP server vetting and free domain checks. Read-only, no acc

InstaVision

Find Instagram creators and leads by niche, city, follower range or lookalike accounts.

3gpp-mcp

higebu

An MCP (Model Context Protocol) server that makes 3GPP specifications accessible to LLMs.

Free SEO Audit for Claude

houtini.com

Free SEO Audit MCP server for Claude - captures and stores all the raw data you need for a complete technical SEO audit. Google Search Console + a first-party crawl + DataForSEO + Majestic SEO merged into one prioritised