CamoFox

CamoFox MCP wraps the CamoFox anti-detection browser as an MCP server, enabling your AI agents to navigate, interact, and automate the web with reduced detection risk and enhanced session persistence.

How to use

You connect an MCP client to the CamoFox MCP server to control browser tabs, navigate pages, read the accessibility tree, and perform actions like clicking, typing, and form submission. Each tab receives a unique fingerprint to help you avoid detection, while built-in search macros and cookie import support streamline ongoing sessions.

How to install

Prerequisites: ensure you have Node.js installed on your system. You will run either a prebuilt MCP client configuration that uses npx, or start the MCP server directly from source for development.

Additional setup steps

{
  "servers": {
    "camofox": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "camofox-mcp@latest"],
      "env": {
        "CAMOFOX_URL": "http://localhost:9377"
      }
    }
  }
}

Notes on development and running from source

If you are developing locally, build and run the MCP server from source, then connect using the local node entry point.

git clone https://github.com/redf0x1/camofox-mcp.git
cd camofox-mcp
npm install
npm run build

{
  "servers": {
    "camofox_dev": {
      "type": "stdio",
      "command": "node",
      "args": ["/path/to/camofox-mcp/dist/index.js"],
      "env": {
        "CAMOFOX_URL": "http://localhost:9377"
      }
    }
  }
}

Environment variables you may use

These variables configure how the MCP client talks to the CamoFox MCP server. Provide them in your MCP client configuration as needed.

Configuration and environment

The following environment variables are shown as part of the setup.

Security and usage notes

Rely on per-tab fingerprints and anti-detection features to minimize automated behavior being flagged, while preserving session continuity and the ability to search across multiple engines.

Tools and capabilities overview

The server exposes a broad set of tools to manage tabs, navigate pages, interact with elements, perform batch actions, inspect page structure, and handle sessions. You can create and close tabs, navigate, click elements, type text, scroll, and take snapshots of the page’s accessibility tree or a visual screenshot.

Architecture

AI Agent → MCP Protocol (stdio) → CamoFox MCP → REST API → CamoFox Browser → Camoufox engine.

Anti-detection features

Each tab receives a unique, human-like fingerprint including user agent rotation, WebGL and canvas fingerprint protections, screen resolution randomization, font enumeration protection, and masking of navigator properties to resemble true user behavior.

Contributing

Contributions are welcome. Open issues or submit pull requests to improve functionality and compatibility with various agents.

License

MIT

Acknowledgments

CamoFox Browser, Camoufox, and the Model Context Protocol underpin the MCP server’s capabilities.

Available tools

create_tab

Create a new tab with anti-detection fingerprinting

close_tab

Close a tab and release resources

list_tabs

List all open tabs with URLs and titles

navigate

Navigate to a URL and wait for the page to load

go_back

Navigate back in history

go_forward

Navigate forward in history

refresh

Reload the current page

click

Click an element by reference or CSS selector

type_text

Type text into input fields by reference or CSS selector

press_key

Press keyboard keys such as Enter, Tab, Escape

scroll

Scroll the page by a specified amount of pixels

fill_form

Fill multiple form fields in one call with optional submission

type_and_submit

Type into a field and submit by key (default Enter)

navigate_and_snapshot

Navigate to a URL, wait for readiness, and return a snapshot

scroll_and_snapshot

Scroll then capture a fresh snapshot

batch_click

Click multiple elements sequentially and return per-click results

snapshot

Get accessibility tree for reading page structure in a token-efficient way

screenshot

Capture a visual screenshot in base64 PNG format

get_links

Extract all hyperlinks with URLs and link text

web_search

Perform searches across 14 engines such as Google, YouTube, and more

import_cookies

Import cookies to maintain authenticated sessions

get_stats

Retrieve session statistics and performance metrics

server_status

Check server health and connection status