Brave Search

An MCP server for integrating with the Brave Search API, and it supports HTTP proxying.
  • javascript

2

GitHub Stars

javascript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "kwp-lab-mcp-brave-search": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "BRAVE_API_KEY",
        "mcp/brave-search"
      ],
      "env": {
        "BRAVE_API_KEY": "YOUR_API_KEY_HERE",
        "BRAVE_SEARCH_PROXY": "https://example.com:10890"
      }
    }
  }
}

You can run the Brave Search MCP Server to perform web and local searches through a unified MCP interface. It leverages the Brave Search API, supports optional proxy routing, and exposes web and local search capabilities with sensible defaults and fallbacks for local results.

How to use

You interact with the Brave Search MCP Server through an MCP client. The server exposes two primary tools: a web search endpoint for queries, and a local search endpoint for nearby businesses and services. You can control search behavior with inputs such as the query string, result count, and offset, and you can direct requests through a proxy if your network requires it.

Practical usage patterns include performing general web searches with pagination, retrieving local business results, and relying on automatic fallback from local to web results when no local matches are found. You can provide your Brave Search API key as an environment variable when starting the server, and optionally configure a proxy URL if needed.

How to install

Prerequisites: you need a system with Docker installed, or you can run the MCP server via NPX. Have your Brave Search API key ready to configure the server securely.

Choose one of the installation methods below and follow the steps exactly as shown.

Docker

{
  "mcpServers": {
    "brave-search": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "BRAVE_API_KEY",
        "mcp/brave-search"
      ],
      "env": {
        "BRAVE_API_KEY": "YOUR_API_KEY_HERE",
        "BRAVE_SEARCH_PROXY": "https://example.com:10890" // Optional, remove if not needed
      }
    }
  }
}

NPX

{
  "mcpServers": {
    "brave-search": {
      "command": "npx",
      "args": [
        "-y",
        "@kwp-lab/mcp-brave-search"
      ],
      "env": {
        "BRAVE_API_KEY": "YOUR_API_KEY_HERE",
        "BRAVE_SEARCH_PROXY": "https://example.com:10890" // Optional, remove if not needed
      }
    }
  }
}

Usage with VS Code and manual setup

For quick setup, you can configure the MCP client in your editor by adding the appropriate mcpServers entry to your user settings. Use either the Docker-based or NPX-based configuration shown above. When you are ready, replace YOUR_API_KEY_HERE with your Brave Search API key and adjust the proxy value if you use a proxy.

Additional notes

The server automatically uses the http_proxy and https_proxy environment variables if they are set, routing requests through the specified proxy. You may also set BRAVE_SEARCH_PROXY to override the proxy for Brave Search requests.

If you plan to run locally, ensure you provide your Brave API key securely and do not commit it to any shared configuration. Use environment-variable-based configuration to keep secrets out of source files.

Available tools

brave_web_search

Execute web searches with pagination and filtering. Input includes query, count per page, and offset for results.

brave_local_search

Search for local businesses and services. Automatically falls back to web search if no local results are found.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational