Camoufox

Provides browser automation via Camoufox with fingerprint spoofing, privacy options, and screenshot capabilities.

typescript

9

GitHub Stars

typescript

Language

5 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": {
    "whit3rabbit-camoufox-mcp": {
      "command": "npx",
      "args": [
        "camoufox-mcp-server"
      ]
    }
  }
}

You can run Camoufox MCP Server to automate browser actions with Camoufox, a privacy-focused browser that helps you browse and capture content while reducing fingerprinting. This MCP server exposes a browse tool your AI assistant can use to navigate websites, take screenshots, and retrieve HTML content with configurable privacy and performance options.

How to use

You will use the browse tool from your MCP client to navigate to websites, fetch content, and capture screenshots when needed. Start by ensuring the server is running via one of the supported methods, then invoke the browse tool with the URL and optional parameters to tailor the browsing session. The tool supports options like OS spoofing, wait strategies, viewport size, timeouts, and privacy controls. Use it to perform tasks such as searching for information, visiting pages, or scraping content, while optionally blocking images or WebRTC and routing through a proxy for privacy.

Basic usage patterns

Open a page and fetch its HTML content for analysis, then optionally capture a screenshot for visual reference.

Advanced usage cues

Combine privacy features with performance tweaks to speed up tasks or protect your identity while browsing sensitive sites.

Example scenarios

Visit example.com with a standard window, wait for the page to finish loading, and capture a screenshot.

Available tools

browse

The browse tool navigates to a given URL, optionally spoofing the OS, applying a wait strategy, setting a viewport, blocking media, routing through a proxy, taking screenshots, and returning page content and optional screenshots for downstream processing.