Tavily Search MCP Server

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

Search tools Artificial intelligence chatbots #Search Optimization #LLM Tools #API Integration Remote .TypeScript

rating : 2.5 points

downloads : 9.0K

update time : 2025-04-29

Open Site

What is the Tavily Search MCP Server?

The Tavily Search MCP Server is a tool that integrates the Tavily Search API, designed specifically for large language models (LLMs). It helps you quickly obtain high-quality web information by optimizing search depth, topics, and time ranges.

How to use the Tavily Search MCP Server?

Using the server is very simple. You can complete the installation and configuration in just a few steps. You can easily call its functions through the command-line tool or the graphical interface.

Application Scenarios

Suitable for application scenarios that require efficient web search, content extraction, and optimization support for LLMs, such as knowledge Q&A and data analysis.

Main Features

Web Search

Perform web searches optimized for LLMs, supporting multiple filtering conditions and options.

Content Extraction

Automatically extract the most relevant paragraphs from the search results to ensure the content quality and appropriate size.

Additional Features

Includes features such as image, image description, short answer generation, and raw HTML content extraction.

Domain Filtering

Supports result filtering that includes or excludes specific domains.

Advantages

Powerful web search capabilities, supporting multiple advanced filtering conditions.

Optimized search results are more suitable for LLMs to process.

Easy to integrate into existing systems without complex settings.

Limitations

Requires a valid Tavily API key to run.

Some advanced features may depend on the subscription plan.

How to Use

Clone the Code Repository

Clone the GitHub repository of this project on your local computer.

Install Dependencies

Install the required dependency packages after entering the project directory.

Build the Project

You can start the server after the build is completed.

Configure Claude Desktop

Add the server to the Claude Desktop configuration file.

Usage Examples

Case Title: Search for Articles Related to Artificial Intelligence

Search for the latest news about artificial intelligence and extract the key paragraphs from them.

Case Title: Extract Content from a Specific Website

Extract information about machine learning from the specified website and generate a brief summary.

Frequently Asked Questions

How to obtain a Tavily API key?

Does it support the HTTPS protocol?

How to stop the Docker container?

Related Resources

Official Documentation

Learn the detailed instructions of the Tavily API.

GitHub Repository

View the source code and contribution guidelines.

Claude Desktop

Download and install the Claude Desktop client.

🚀 Tavily Search MCP Server

A plugin designed for Claude Desktop, offering powerful search capabilities through the Tavily API. Users can utilize the service by configuring the environment variable TAVILY_API_KEY.

🚀 Quick Start

The Tavily Search MCP Server is a plugin crafted for Claude Desktop. It leverages the Tavily API to deliver robust search functionality. You can start using the service by setting up the TAVILY_API_KEY environment variable.

✨ Features

Multilingual Support: Supports multiple programming languages.
Scalability: Allows for custom extensions based on requirements.
High Availability: Ensures stable operation under high loads.
Security: Manages API keys securely via environment variables to prevent sensitive information leakage.

📦 Installation

Automatic Installation via Smithery

Quickly install through Smithery:

npx -y @smithery/cli install @apappascs/tavily-search-mcp-server --client claude

📦 Environment Configuration (for npm)

Copy the .env.example file and rename it to .env:
```
cp .env.example .env
```
Fill in the actual Tavily API key in the .env file:
```
TAVILY_API_KEY=your_api_key_here
```
⚠️ Important Note

Do not commit the actual key to the version control system. The .env file is ignored by git for security reasons.

🚀 Running the Server

Starting with npm

Standard Input/Output Mode:
```
node dist/index.js
```
SSE (Server-Sent Events) Transmission Mode:
```
node dist/sse.js
```

Running with Docker

Build the Docker image (if not already built):

docker build -t tavily-search-mcp-server:latest .

Start the container:

Standard Input/Output Mode:

docker run -it --rm -e TAVILY_API_KEY="your_api_key_here" tavily-search-mcp-server:latest

SSE Transmission Mode:

docker run -it --rm -p 3001:3001 -e TAVILY_API_KEY="your_api_key_here" -e TRANSPORT="sse" tavily-search-mcp-server:latest

💡 Usage Tip

It is recommended to use environment variables to pass the API key to avoid exposing sensitive information directly in the command line:

docker run -it --rm -p 3001:3001 -e TAVILY_API_KEY=$TAVILY_API_KEY -e TRANSPORT="sse" tavily-search-mcp-server:latest

Managing with Docker Compose

To start the service:

docker compose up -d

To stop the service:

docker compose down

📄 License

This MCP server is licensed under the MIT License. This means you are free to use, modify, and distribute the software, subject to the terms and conditions of the MIT License. For more details, please refer to the LICENSE file in the project repository.