Brave Search MCP Server

What is this?

The Brave Search MCP Server is a specialized Model Context Protocol server that lets your AI agents perform real-time web searches using the Brave Search API. It exposes an MCP-compliant interface so that large language models can fetch up-to-date information from the web instead of relying solely on static training data. This makes your AI responses more accurate, relevant, and timely.

Under the hood, the server handles HTTP requests following the MCP REST protocol, translates them into Brave Search API calls, and returns structured MCP-compatible responses. Whether you’re building a fact-checking assistant, a market-monitoring tool, or a dynamic content generator, this server provides a plug-and-play solution to integrate live web search with minimal setup.

Quick Start

Install the server using npm:

npm install @modelcontextprotocol/server-brave-search

Then add it to your MCP client configuration:

{
  "mcpServers": {
    "brave-search": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-brave-search"],
      "env": {
        "API_KEY": "your-api-key-here"
      }
    }
  }
}

Key Features

Feature 1: Real-time Web Search – Submit natural-language queries to Brave Search API and get immediate, up-to-date results for news, market data, and more.

Feature 2: Privacy-Centric Search – Built on Brave’s privacy-first architecture, the server never tracks user queries or personal data.

Feature 3: Seamless MCP Compliance – Out-of-the-box adherence to the MCP specification, making it simple to plug into existing AI frameworks.

Example Usage

Here’s how you might use the server to add live search capabilities to your AI assistant. You can call the search tool directly from your MCP-compatible client to retrieve fresh web results:

// Example code
const result = await client.callTool({
  name: "search",
  arguments: {
    query: "latest electric vehicle market trends",
    pageSize: 5,
    region: "us"
  }
});

This code sends a query to the Brave Search MCP Server and returns an array of search results, each with a title, snippet, and URL.

Configuration

The server accepts the following environment variables:

API_KEY – Your Brave Search API key, which you can obtain by signing up at https://search.brave.com/account.

DEFAULT_REGION (optional) – The default two-letter region code for searches (e.g., us, gb).

Available Tools/Resources

search: A tool that wraps the /v1/search endpoint to submit queries and receive structured MCP-compatible results.

/v1/search: The primary REST endpoint for performing web searches via the Brave Search MCP Server.

Who Should Use This?

This server is perfect for:

Use case 1: Journalistic AI assistants that need to verify facts with the latest news sources.

Use case 2: Market monitoring tools tracking real-time financial data and trends.

Use case 3: Dynamic content generators that include up-to-the-minute information in their outputs.

Conclusion

The Brave Search MCP Server lets you bridge your AI agents to live web data in a privacy-focused, MCP-compliant way. With easy setup, real-time search capabilities, and seamless integration, it’s the go-to solution for adding fresh information to any AI workflow.

Check out the GitHub repository for more information and to contribute.