Brave Deep Research

Provides deep web research by combining Brave Search results with page content extraction and link traversal.

typescript

GitHub Stars

typescript

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": {
    "suthio-brave-deep-research-mcp": {
      "command": "npx",
      "args": [
        "@suthio/brave-deep-research-mcp"
      ],
      "env": {
        "DEBUG_MODE": "false",
        "PAGE_TIMEOUT": "30000",
        "BRAVE_API_KEY": "YOUR_BRAVE_API_KEY",
        "PUPPETEER_HEADLESS": "true"
      }
    }
  }
}

You run Brave Deep Research MCP Server to perform in-depth web research. It starts from Brave Search results, then visits pages to extract full content, follow links, and present richer context for your AI assistants. This makes it ideal when you need comprehensive information and connections across related pages.

How to use

Start the server and connect your MCP client to perform deep research queries. You provide a search query, the server fetches initial results from Brave Search, visits each result page, extracts main content, and can follow linked pages up to a configurable depth. Use this to gather detailed information, verify sources, and build context across related topics.

How to install

Prerequisites you need before installation:

Node.js and npm installed on your system
Git installed (optional for cloning)

Install or clone the server, install dependencies, and build before first run.

Configuration

Set up environment variables for Brave API access and runtime behavior. Create a local environment file and populate the required values.

Usage examples

# If you install globally via npm
brave-deep-research-mcp

# Or run directly from the package
npx @suthio/brave-deep-research-mcp

# Or run locally after cloning
npm start

Advanced configuration and troubleshooting

Configure how deep the server follows links, how many results to process, and how the browser runs (headless or visible). If you run into timeouts or debugging needs, adjust timeout and debug mode settings.

Example configuration for client integration

{
  "mcpServers": {
    "brave_deep_research": {
      "command": "npx",
      "args": ["@suthio/brave-deep-research-mcp"],
      "env": {
        "BRAVE_API_KEY": "YOUR_BRAVE_API_KEY",
        "PUPPETEER_HEADLESS": "true"
      }
    }
  }
}

Tools and capabilities

The server provides a Deep Search feature set, including full content extraction, configurable link-depth traversal, metadata extraction, and adjustable runtime options such as headless mode and timeouts.

Available tools

deepSearch

Performs a Brave Search query and processes results through deep content extraction and link traversal

contentExtraction

Extracts main textual content from visited pages, filtering out navigation, ads, and footers

metadataExtraction

Retrieves titles, descriptions, and structured content from pages and linked pages

depthTraversal

Configures how many levels of links to follow from each initial result and repeats the extraction on those pages

debugAndRuntime

Supports enabling detailed logs and toggling Puppeteer headless mode for troubleshooting

Built by

VeilStrat

AI signals for GTM teams