Installation Guide - JinsongRoh/pydoll-mcp GitHub Wiki
This comprehensive guide walks you through installing PyDoll MCP Server on Windows, macOS, and Linux systems.
- Python: 3.8 or higher (Recommended: 3.9+)
- OS: Windows 10+, macOS 10.14+, Linux (Ubuntu 18.04+, CentOS 7+)
- Browser: Chrome or Edge (auto-detected)
- Memory: Minimum 4GB RAM (Recommended: 8GB)
- Storage: 500MB free space
- Python 3.8+ installed
- pip package manager available
- Chrome or Edge browser installed
- Internet connection (for package downloads)
- Administrator/sudo privileges (for system-wide installation)
# Run Command Prompt or PowerShell as Administrator
# Install PyDoll MCP Server
pip install pydoll-mcp
# Verify installation
pydoll-mcp --version
python -m pydoll_mcp.server --test
# Open Terminal
# Install PyDoll MCP Server
pip install pydoll-mcp
# If permission issues occur:
pip install --user pydoll-mcp
# Verify installation
pydoll-mcp --version
python -m pydoll_mcp.server --test
# Update system packages
sudo apt update
# Install Python pip (if needed)
sudo apt install python3-pip
# Install PyDoll MCP Server
pip3 install pydoll-mcp
# Verify installation
pydoll-mcp --version
python3 -m pydoll_mcp.server --test
# Enable EPEL repository
sudo yum install epel-release
# Install Python pip
sudo yum install python3-pip
# Install PyDoll MCP Server
pip3 install pydoll-mcp
# Verify installation
pydoll-mcp --version
python3 -m pydoll_mcp.server --test
# Clone repository
git clone https://github.com/JinsongRoh/pydoll-mcp.git
cd pydoll-mcp
# Create virtual environment (recommended)
python -m venv venv
# Activate virtual environment
# Windows:
venv\Scripts\activate
# Linux/macOS:
source venv/bin/activate
# Install dependencies
pip install -r requirements.txt
# Install in development mode
pip install -e .
# Download and run Docker image
docker run -d --name pydoll-mcp -p 8080:8080 jinsongroh/pydoll-mcp:latest
# Check container status
docker ps
# View container logs
docker logs pydoll-mcp
# docker-compose.yml
version: '3.8'
services:
pydoll-mcp:
image: jinsongroh/pydoll-mcp:latest
ports:
- "8080:8080"
environment:
- PYDOLL_LOG_LEVEL=INFO
volumes:
- ./config:/app/config
# Start service
docker-compose up -d
# Check status
docker-compose ps
# Run auto setup after installation
python -m pydoll_mcp.cli auto-setup
This command performs the following:
- โ Test installation status
- โ Verify Claude Desktop configuration file location
- โ Backup existing configuration
- โ Add PyDoll MCP Server configuration
- โ Verify configuration works
# Step-by-step setup guide
python -m pydoll_mcp.cli quick-start
-
Windows:
%APPDATA%\Claude\claude_desktop_config.json
-
macOS:
~/Library/Application Support/Claude/claude_desktop_config.json
-
Linux:
~/.config/Claude/claude_desktop_config.json
{
"mcpServers": {
"pydoll": {
"command": "python",
"args": ["-m", "pydoll_mcp.server"],
"env": {
"PYDOLL_LOG_LEVEL": "INFO",
"PYTHONIOENCODING": "utf-8"
}
}
}
}
# Auto-generate configuration file
python -m pydoll_mcp.cli generate-config
# Generate in various formats
python -m pydoll_mcp.cli generate-config --format yaml
python -m pydoll_mcp.cli generate-config --format env
# Download and run setup script
curl -o setup_claude.bat https://raw.githubusercontent.com/JinsongRoh/pydoll-mcp/main/setup/setup_claude_windows.bat
setup_claude.bat
# Download and run setup script
curl -o setup_claude.sh https://raw.githubusercontent.com/JinsongRoh/pydoll-mcp/main/setup/setup_claude_unix.sh
chmod +x setup_claude.sh
./setup_claude.sh
# Set environment variables
set PYDOLL_LOG_LEVEL=INFO
set PYTHONIOENCODING=utf-8
# Permanent setting
setx PYDOLL_LOG_LEVEL "INFO"
setx PYTHONIOENCODING "utf-8"
# Set for current session
export PYDOLL_LOG_LEVEL=INFO
export PYTHONIOENCODING=utf-8
# Permanent setting (add to ~/.bashrc or ~/.zshrc)
echo 'export PYDOLL_LOG_LEVEL=INFO' >> ~/.bashrc
echo 'export PYTHONIOENCODING=utf-8' >> ~/.bashrc
# Complete installation test
python -m pydoll_mcp.cli test-installation --verbose
# Browser test
python -m pydoll_mcp.cli test-browser --browser chrome --headless
# Check server status
python -m pydoll_mcp.cli status --logs --stats
# Check Python version
python --version
# Verify package installation
pip show pydoll-mcp
# Verify browser detection
python -c "from pydoll.browser import Chrome; print('Chrome detection successful')"
# Test MCP server
python -m pydoll_mcp.server --test
# Check Python version
python --version
# Recommended: Use Python 3.9 or higher
# Upgrade method varies by OS
# Upgrade pip
python -m pip install --upgrade pip
# Detailed installation log
pip install pydoll-mcp -v
# User installation (Linux/macOS)
pip install --user pydoll-mcp
# Administrator privileges (Windows)
# Run PowerShell as Administrator
# Increase timeout
pip install --timeout 60 pydoll-mcp
# Use different index
pip install -i https://pypi.org/simple/ pydoll-mcp
# Windows
"C:\Program Files\Google\Chrome\Application\chrome.exe" --version
# macOS
/Applications/Google\ Chrome.app/Contents/MacOS/Google\ Chrome --version
# Linux
google-chrome --version
# Windows
"C:\Program Files (x86)\Microsoft\Edge\Application\msedge.exe" --version
# macOS
/Applications/Microsoft\ Edge.app/Contents/MacOS/Microsoft\ Edge --version
# Linux
microsoft-edge --version
# Set UTF-8 in Command Prompt
chcp 65001
# Set environment variable
set PYTHONIOENCODING=utf-8
# Add to Claude Desktop configuration
{
"mcpServers": {
"pydoll": {
"command": "python",
"args": ["-m", "pydoll_mcp.server"],
"env": {
"PYTHONIOENCODING": "utf-8",
"PYDOLL_LOG_LEVEL": "INFO"
}
}
}
}
# Upgrade to latest version
pip install --upgrade pydoll-mcp
# Verify fixes
python -m pydoll_mcp.server --test
# Check server output (should be JSON format)
python -m pydoll_mcp.cli status
# Install with development dependencies
pip install "pydoll-mcp[dev]"
# Install test tools
pip install "pydoll-mcp[test]"
# Install all dependencies
pip install "pydoll-mcp[dev,test]"
# Create venv
python -m venv pydoll-env
# Activate
# Windows:
pydoll-env\Scripts\activate
# Linux/macOS:
source pydoll-env/bin/activate
# Install
pip install pydoll-mcp
# Create Conda environment
conda create -n pydoll-env python=3.9
# Activate environment
conda activate pydoll-env
# Install
pip install pydoll-mcp
Follow the [โก Quick Start Guide](Quick-Start) to run your first automation.
Check [๐ฏ Basic Usage](Basic-Usage) for basic commands and examples.
See [๐ ๏ธ Tools Reference](Tools-Reference) for all available tools.
Learn performance optimization methods in [โ๏ธ Advanced Configuration](Advanced-Configuration).
- Use Virtual Environments: Isolated environments per project
- Use Latest Python: 3.9 or higher recommended
-
Regular Updates:
pip install --upgrade pydoll-mcp
- Check System Resources: Ensure sufficient memory and storage
- Trusted Sources: Use official PyPI repository
- Permission Management: Grant only minimum necessary permissions
- Firewall Settings: Open only required ports
- Log Monitoring: Regular log checking
Installation Complete! ๐
You're now ready to use PyDoll MCP Server.