camoufox -python
MCP server for browser automation with Camoufox anti-detection capabilities
claude mcp add --transport stdio rlgrpe-camoufox-mcp-python uvx camoufox-mcp
How to use
The Camoufox MCP server provides a Python-based Model Context Protocol (MCP) interface to automate a Camoufox-powered browser with anti-detection features. It ships with a suite of 21 tools organized into Navigation, Interaction, Page, Forms, and Session categories, enabling complete browser control, form automation, and tab management while preserving sessions across calls. Typical usage includes starting the MCP server, then issuing tool invocations such as browser_navigate, browser_click, browser_type, browser_take_screenshot, and browser_tabs to manage tabs and sessions. The server lazily initializes the browser on the first tool call and maintains the browser state between tool invocations, helping you script complex interactions across multiple steps without re-launching the browser each time. When integrated with Claude or other orchestration layers, you can add the MCP server as a resource and call its tools to perform web automation tasks while benefiting from anti-detection capabilities during automated browsing.
How to install
Prerequisites:
- Python 3.8+ installed
- Your preferred method to run Python packages (uv, pipx, or pip)
- Git (if you plan to install from source)
Installation (recommended quick start using uvx):
- Install and run the MCP server via uvx:
uvx camoufox-mcp
- Alternatively, install with uv (from source or local development):
uv run camoufox-mcp
3) Install via pipx (recommended for isolated environments):
```bash
pipx run camoufox-mcp
4) Install via pip (if you prefer direct Python package installation):
```bash
pip install camoufox-mcp
camoufox-mcp
5) If building from source:
```bash
# Clone the repository
git clone https://github.com/rlgrpe/camoufox-mcp-python.git
cd camoufox-mcp-python
# Install with uv
uv sync
- Install Camoufox browser binary after package installation:
- If using uvx (recommended):
uvx camoufox fetch
- If using pipx:
pipx run camoufox fetch
- If using pip:
# macOS
python3 -m camoufox fetch
# Linux
python -m camoufox fetch
# Windows
camoufox fetch
- On Linux, install additional system dependencies as needed:
# Ubuntu/Debian
sudo apt install -y libgtk-3-0 libx11-xcb1 libasound2
# Arch
sudo pacman -S gtk3 libx11 libxcb cairo libasound alsa-lib
- To uninstall the browser binary, use the same method as installation:
# uvx
uvx camoufox remove
# pipx
pipx run camoufox remove
# pip
camoufox remove
# from source (uv)
uv run python -m camoufox remove
Additional notes
Tips and considerations:
- The browser initializes lazily on the first tool call, which can help speed up startup in scripted scenarios.
- Persistent sessions allow multi-step automation without re-creating the browser context between calls.
- If you encounter environment-specific issues, ensure the Camoufox binary has proper permissions and the system dependencies for your OS are installed.
- The Tools Reference in the README enumerates 21 tools across Navigation, Interaction, Page, Forms, and Session categories; use them to build end-to-end automation flows.
- When integrating with Claude or other tooling, you can add the server under mcpServers with the uvx camoufox-mcp command, or adjust to the appropriate runtime if running from source.
- If you need to customize browser behavior, consult the BrowserConfig options in camoufox_mcp/config.py, including headless mode, geolocation, proxy, and fingerprint randomization controls.
Related MCP Servers
mcp -qdrant
An official Qdrant Model Context Protocol (MCP) server implementation
mcp-pinecone
Model Context Protocol server to allow for reading and writing from Pinecone. Rudimentary RAG
pfsense
pfSense MCP Server enables security administrators to manage their pfSense firewalls using natural language through AI assistants like Claude Desktop. Simply ask "Show me blocked IPs" or "Run a PCI compliance check" instead of navigating complex interfaces. Supports REST/XML-RPC/SSH connections, and includes built-in complian
mcp-kubernetes
A Model Context Protocol (MCP) server that enables AI assistants to interact with Kubernetes clusters. It serves as a bridge between AI tools (like Claude, Cursor, and GitHub Copilot) and Kubernetes, translating natural language requests into Kubernetes operations and returning the results in a format the AI tools can understand.
davinci -professional
An enterprise-grade MCP server that exposes the full functionality of DaVinci Resolve and DaVinci Resolve Studio (through version 20) to either Claude Desktop or Cursor MCP clients. Fully configured and tested as a Claude Desktop Extension making installation as easy as clicking a button. Supports both Windows and Macintosh.
mycobot
myCobot MCP(Model Context Protocol) server