🤖 PyDoll MCP Server FAQ

Find answers to frequently asked questions. For more questions, please ask in [Issues](https://github.com/JinsongRoh/pydoll-mcp/issues).

📋 Table of Contents

🏗️ Installation & Setup

[What is PyDoll MCP Server?](#what-is-pydoll-mcp)
[How to install?](#how-to-install)
[How to integrate with Claude Desktop?](#claude-integration)
[Auto setup doesn't work after installation](#auto-setup-issues)
[How to set up manually?](#manual-setup)

🚨 Troubleshooting

[JSON parsing error occurs](#json-parsing-error)
[Encoding error on Korean Windows](#korean-encoding-error)
[Browser doesn't start](#browser-not-starting)
[Captcha bypass doesn't work](#captcha-bypass-not-working)
[Network connection error occurs](#network-connection-error)

🏗️ Installation & Setup

What is PyDoll MCP Server? {#what-is-pydoll-mcp}

Answer: PyDoll MCP Server is a revolutionary browser automation tool:

🚫 Zero WebDriver: Direct browser communication via Chrome DevTools Protocol
🧠 AI-powered captcha bypass: Automatic solving of Cloudflare Turnstile & reCAPTCHA v3
👤 Human behavior simulation: Undetectable interactions that fool sophisticated anti-bot systems
⚡ Native async architecture: Lightning-fast concurrent automation
🕵️ Advanced stealth mode: Anti-detection techniques that make automation invisible

Compatible with AI clients that support MCP, including Claude, OpenAI GPT, and Gemini.

How to install? {#how-to-install}

Answer: Installation is very simple:

🚀 Recommended method (with auto setup)

pip install pydoll-mcp

After installation, auto setup options will appear automatically:

🤖 Setting up PyDoll MCP Server...

🎉 PyDoll MCP Server installed successfully!
============================================================

🚀 Quick Start Options:
1. 🔧 Auto-configure Claude Desktop
2. 📋 Generate config manually  
3. 🧪 Test installation
4. ⏭️  Skip setup (configure later)

🎯 Choose an option (1-4): 1

🔧 Additional setup commands

# One-click auto setup
python -m pydoll_mcp.cli auto-setup

# Setup Claude Desktop only
python -m pydoll_mcp.cli setup-claude

# Interactive guide
python -m pydoll_mcp.cli quick-start

How to integrate with Claude Desktop? {#claude-integration}

Answer: Multiple ways to integrate:

🔧 Auto setup (easiest method)

python -m pydoll_mcp.cli auto-setup

This command automatically:

✅ Tests installation
✅ Finds Claude Desktop config file location
✅ Backs up existing configuration
✅ Adds PyDoll MCP Server configuration
✅ Verifies everything works properly

🛠️ Manual setup

Config file locations:

Windows: %APPDATA%\Claude\claude_desktop_config.json
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Linux: ~/.config/Claude/claude_desktop_config.json

Configuration content:

{
  "mcpServers": {
    "pydoll": {
      "command": "python",
      "args": ["-m", "pydoll_mcp.server"],
      "env": {
        "PYDOLL_LOG_LEVEL": "INFO"
      }
    }
  }
}

Auto setup doesn't work after installation {#auto-setup-issues}

Answer: Try the following steps:

Check permissions

# Check if Claude Desktop is installed
python -m pydoll_mcp.cli status

# Manually generate config
python -m pydoll_mcp.cli generate-config --auto-setup

Force reset

python -m pydoll_mcp.cli auto-setup --force

Check config file directly

# Windows
type "%APPDATA%\Claude\claude_desktop_config.json"

# macOS/Linux
cat ~/Library/Application\ Support/Claude/claude_desktop_config.json

Restart Claude Desktop
- Completely close Claude Desktop
- Restart
- Check if "PyDoll" tool appears

How to set up manually? {#manual-setup}

Answer: Manual setup guide:

Create config file

python -m pydoll_mcp.cli generate-config

Edit config file
- Windows: Open %APPDATA%\Claude\claude_desktop_config.json with Notepad
- macOS/Linux: Open config file with text editor
Add/modify content

{
  "mcpServers": {
    "pydoll": {
      "command": "python",
      "args": ["-m", "pydoll_mcp.server"],
      "env": {
        "PYDOLL_LOG_LEVEL": "INFO",
        "PYTHONIOENCODING": "utf-8"
      }
    }
  }
}

Validate configuration

python -m pydoll_mcp.cli test-installation --verbose

🚨 Troubleshooting

JSON parsing error occurs {#json-parsing-error}

Answer: This issue has been fixed in v1.1.3!

🔧 Solution:

Upgrade to latest version

pip install --upgrade pydoll-mcp

Check version

python -c "import pydoll_mcp; print(pydoll_mcp.__version__)"
# Should be 1.1.3 or higher

Verify fixes

python -m pydoll_mcp.server --test
# Should output clean JSON

💡 v1.1.3 fixes:

stdout/stderr separation: Moved banner output to stderr for MCP protocol compliance
JSON validity: Ensured all stdout output is valid JSON
Improved error handling: Better error messages and JSON formatting

Encoding error on Korean Windows {#korean-encoding-error}

Answer: Korean Windows support has been greatly improved in v1.1.3!

🔧 Immediate solutions:

Set environment variables

set PYTHONIOENCODING=utf-8
python -m pydoll_mcp.server

Change code page

chcp 65001
python -m pydoll_mcp.server

Add encoding environment variable to Claude Desktop config

{
  "mcpServers": {
    "pydoll": {
      "command": "python",
      "args": ["-m", "pydoll_mcp.server"],
      "env": {
        "PYTHONIOENCODING": "utf-8",
        "PYDOLL_LOG_LEVEL": "INFO"
      }
    }
  }
}

💡 Permanent solution:

System locale setting: Windows Settings → Time & Language → Language → Administrative language settings → Use Unicode UTF-8 support

Browser doesn't start {#browser-not-starting}

Answer: Step-by-step troubleshooting:

Check browser installation

# Check if Chrome is installed
python -c "from pydoll.browser import Chrome; print('Browser check passed')"

Test basic functionality

python -m pydoll_mcp.cli test-browser --browser chrome

Resolve permission issues

# Check Chrome permissions on Linux
ls -la /usr/bin/google-chrome

# Try running Chrome directly
google-chrome --version

Try headless mode

python -m pydoll_mcp.cli test-browser --browser chrome --headless

Run in debug mode

export PYDOLL_DEBUG=1
export PYDOLL_LOG_LEVEL=DEBUG
python -m pydoll_mcp.server --debug

🔍 Common causes and solutions:

Chrome not installed → Install Chrome
Insufficient permissions → Run with administrator privileges
Port conflicts → Use different port
Insufficient memory → Check system resources

Captcha bypass doesn't work {#captcha-bypass-not-working}

Answer: Captcha bypass optimization methods:

🧠 Supported captcha types:

Cloudflare Turnstile (95%+ success rate)
reCAPTCHA v3 (90%+ success rate)
Basic image captcha

🔧 Improve bypass performance:

Enable stealth mode

# Ask Claude like this:
"Enable stealth mode and navigate to the protected site"
"Automatically solve any captcha challenges that appear"

Use human behavior simulation

"Enable human behavior simulation before attempting captcha bypass"

Adjust retry settings

{
  "captcha_config": {
    "auto_solve_cloudflare": true,
    "auto_solve_recaptcha": true,
    "solve_timeout": 30,
    "retry_failed_attempts": 3,
    "human_behavior_simulation": true
  }
}

💡 Tips to increase success rate:

Provide sufficient wait time
Use natural mouse movements
Avoid consistent patterns
Randomize browser fingerprints

Network connection error occurs {#network-connection-error}

Answer: Network troubleshooting:

Test connection

python -m pydoll_mcp.cli status --logs --stats

Check firewall settings
- Allow PyDoll MCP Server ports
- Allow Chrome DevTools Protocol port (9222)
Proxy configuration

# Ask Claude to use proxy:
"Start browser with proxy server 127.0.0.1:8080"

Network condition simulation

"Simulate slow 3G network conditions for testing"

🔧 Features & Usage

Which browsers are supported? {#supported-browsers}

Answer: Currently supported browsers:

✅ Fully supported:

Google Chrome (recommended)
Microsoft Edge (Chromium-based)

🔄 Coming soon (v1.2.0):

Mozilla Firefox
Safari (macOS)

🛠️ Browser-specific settings:

# Use Chrome (default)
"Start Chrome browser with stealth mode enabled"

# Use Edge
"Start Microsoft Edge browser with ad blocking enabled"

# Headless mode
"Start browser in headless mode for background automation"

What's the difference between headless and normal mode? {#headless-vs-normal}

Answer: Mode comparison:

🖥️ Normal mode (with GUI):

Advantages:

Real-time progress monitoring
Easy debugging
User interaction possible
Higher captcha bypass success rate

Disadvantages:

Uses more system resources
Requires display
Relatively slower

👻 Headless mode (no GUI):

Advantages:

Fast execution speed
Lower memory usage
Can run in server environment
Suitable for background tasks

Disadvantages:

Some captcha bypass limitations
Difficult debugging
No visual confirmation

🔧 Usage:

# Normal mode
"Start browser and take a screenshot of the homepage"

# Headless mode  
"Start browser in headless mode and extract page data"

What types of captcha can be bypassed? {#captcha-types}

Answer: Supported captcha types and success rates:

🥇 High success rate (95%+):

Cloudflare Turnstile
Cloudflare Challenge Pages
Basic JavaScript Challenge

🥈 Medium success rate (90%+):

reCAPTCHA v3
hCaptcha
Simple Image Captcha

🥉 Limited support (70%+):

Complex image captcha
Audio captcha
Math problem captcha

🚫 Not supported:

reCAPTCHA v2 (Click Challenge)
Complex puzzle captcha
Custom captcha

💡 Optimization methods:

# Enable automatic captcha bypass
"Enable automatic captcha bypass and navigate to protected site"

# Use with stealth mode
"Enable stealth mode and automatically solve captcha challenges"

# Human behavior simulation
"Simulate human behavior and solve captcha with natural timing"

How does network monitoring work? {#network-monitoring}

Answer: Comprehensive network analysis features:

📊 Monitoring capabilities:

Real-time request/response capture
Automatic API call extraction
Header and payload analysis
WebSocket connection tracking
Performance metrics collection

🔧 Usage methods:

# Basic network monitoring
"Enable network monitoring and browse this e-commerce website"

# Automatic API response extraction
"Monitor all API calls and extract product pricing data"

# Filter specific requests
"Monitor only POST requests to the login endpoint"

# Performance analysis
"Analyze page load performance and network timing"

📈 Output information:

Request URL and method
Response status code
Response time
Data size
Header information
JSON/XML payload

How to upload/download files? {#file-operations}

Answer: Various file operations supported:

📤 File upload:

# Single file upload
"Find the file input and upload document.pdf"

# Multiple file upload
"Upload multiple images to the gallery: image1.jpg, image2.png"

# Drag and drop upload
"Drag and drop the file onto the upload area"

📥 File download:

# Click link download
"Click the download link and save the file"

# Right-click download
"Right-click the image and save it to downloads folder"

# Programmatic download
"Download all PDF files from this page programmatically"

🗂️ Supported file formats:

Images: JPG, PNG, GIF, WebP, SVG
Documents: PDF, DOC, DOCX, XLS, XLSX
Archives: ZIP, RAR, 7Z
Media: MP4, MP3, AVI
Code: TXT, CSV, JSON, XML

⚡ Performance & Optimization

How to reduce memory usage? {#reduce-memory-usage}

Answer: Memory optimization methods:

🔧 Browser settings optimization:

# Disable images
"Start browser with images disabled for faster loading"

# Enable ad blocking
"Enable ad blocking to reduce memory usage"

# Use headless mode
"Use headless mode for background automation tasks"

⚙️ Advanced settings:

{
  "browser_config": {
    "headless": true,
    "disable_images": true,
    "disable_css": false,
    "block_ads": true,
    "enable_compression": true,
    "max_concurrent_tabs": 3
  }
}

💡 Memory management tips:

Limit tab count: Use only 3-5 tabs simultaneously
Regular cleanup: Close unused tabs
Cache management: Periodically clear browser cache
Disable extensions: Remove unnecessary extensions

How to improve speed? {#improve-speed}

Answer: Performance optimization strategies:

🚀 Immediately applicable optimizations:

# Headless mode + disable images
"Start browser in headless mode with images disabled"

# Enable network caching
"Enable network caching for repeated requests"

# Block unnecessary resources
"Block ads, trackers, and analytics for faster loading"

⚙️ Network optimization:

{
  "network_config": {
    "timeout": 15,
    "retry_attempts": 2,
    "enable_caching": true,
    "throttle_requests": false,
    "concurrent_connections": 10
  }
}

📊 Performance comparison:

Setting	Default	Optimized	Improvement
Page load	3.2s	1.1s	65% ↑
Memory usage	450MB	180MB	60% ↓
CPU usage	35%	15%	57% ↓

Can I run multiple browsers simultaneously? {#multiple-browsers}

Answer: Multi-browser support:

🔄 Concurrent execution features:

Multiple browser instances running simultaneously
Independent session management
Tab isolation environment
Resource distribution processing

🛠️ Usage methods:

# First browser
"Start first browser for social media automation"

# Second browser (concurrent)
"Start second browser for e-commerce data extraction"

# List browsers
"List all active browser instances with their status"

# Select specific browser
"Switch to browser instance 1 for the next operations"

📊 Recommended concurrent numbers:

System RAM	Recommended Browsers	Tab Limit
8GB	2	3/browser
16GB	4	5/browser
32GB	8	8/browser

How to improve stability? {#improve-stability}

Answer: Stability improvement methods:

🛡️ Error recovery settings:

{
  "stability_config": {
    "auto_retry_failed_operations": true,
    "max_retry_attempts": 3,
    "retry_delay": 2,
    "graceful_error_handling": true,
    "automatic_cleanup": true
  }
}

🔧 Stability optimization:

# Set sufficient wait times
"Wait for page to fully load before proceeding"

# Verify element existence before manipulation
"Wait for the button to be visible and clickable before clicking"

# Check network stability
"Wait for network to be idle before extracting data"

# Automatic recovery on errors
"If an error occurs, automatically retry the operation"

💡 Stability checklist:

✅ Set sufficient wait times
✅ Validate element existence before manipulation
✅ Check network status
✅ Implement error handling
✅ Regular status checks
✅ Automated resource cleanup

🛡️ Security & Evasion

What is stealth mode? {#stealth-mode}

Answer: Advanced detection evasion technology:

🕵️ Stealth mode features:

Browser fingerprint randomization
User agent rotation
WebRTC leak prevention
Canvas fingerprint masking
Timing attack prevention
Automation detection bypass

🔧 Activation methods:

# Basic stealth mode
"Enable stealth mode and navigate to the protected website"

# Advanced stealth settings
"Enable advanced stealth mode with fingerprint randomization"

# Specific detection bypass
"Bypass WebRTC detection and mask canvas fingerprint"

⚙️ Advanced settings:

{
  "stealth_config": {
    "randomize_fingerprint": true,
    "rotate_user_agents": true,
    "humanize_timing": true,
    "evade_webrtc": true,
    "spoof_timezone": true,
    "mask_canvas": true,
    "disable_webgl": false
  }
}

How to bypass bot detection? {#bot-detection-bypass}

Answer: Multi-layer detection bypass system:

🧠 Detection bypass techniques:

Behavior pattern simulation
- Natural mouse movements
- Human-like typing speed
- Realistic page dwell time
Technical bypass
- Direct Chrome DevTools Protocol usage
- WebDriver trace removal
- JavaScript execution context masking
Network level bypass
- Header spoofing
- Request pattern randomization
- Proxy rotation

🎯 Real usage examples:

# Comprehensive detection bypass
"Enable comprehensive bot detection bypass for this banking website"

# Human behavior simulation
"Simulate realistic human browsing patterns with random delays"

# Advanced evasion techniques
"Use advanced evasion techniques for sophisticated protection"

Can I use proxies? {#proxy-support}

Answer: Complete proxy support:

🌐 Supported proxy types:

HTTP proxy
HTTPS proxy
SOCKS4/SOCKS5 proxy
Authenticated proxy

🔧 Proxy configuration:

# HTTP proxy
"Start browser with HTTP proxy 127.0.0.1:8080"

# Authenticated proxy
"Use proxy 127.0.0.1:8080 with username 'user' and password 'pass'"

# Proxy rotation
"Rotate between multiple proxy servers for each request"

# Direct connection without proxy
"Bypass proxy and connect directly for this request"

🔄 Proxy rotation settings:

{
  "proxy_config": {
    "enabled": true,
    "rotation": true,
    "proxies": [
      "http://proxy1:8080",
      "http://proxy2:8080", 
      "socks5://proxy3:1080"
    ],
    "auth": {
      "username": "user",
      "password": "pass"
    }
  }
}

Can I change user agents? {#user-agent-rotation}

Answer: Dynamic user agent management:

🔄 Automatic rotation:

# Automatic user agent rotation
"Enable automatic user agent rotation for each request"

# Specific OS simulation
"Use Windows 10 Chrome user agents only"

# Mobile device simulation
"Rotate between different mobile device user agents"

📱 Device-specific user agents:

Desktop: Windows, macOS, Linux Chrome/Edge
Mobile: Android, iOS Safari/Chrome
Tablet: iPad, Android Tablet
Custom: User-defined strings

⚙️ Rotation settings:

{
  "user_agent_config": {
    "rotation_enabled": true,
    "rotation_interval": 50,
    "device_types": ["desktop", "mobile"],
    "custom_agents": [
      "Mozilla/5.0 (Windows NT 10.0; Win64; x64)...",
      "Mozilla/5.0 (iPhone; CPU iPhone OS 14_7_1)..."
    ]
  }
}

🔄 Updates & Compatibility

What are the major changes in v1.1.3? {#v113-changes}

Answer: v1.1.3 major improvements:

🐛 Core bug fixes:

✅ JSON parsing error fixed: Resolved MCP client communication issues
✅ Encoding compatibility: Full support for Korean Windows systems (CP949/EUC-KR)
✅ Protocol compliance: Proper stdout/stderr separation for MCP compatibility
✅ Stability improvements: Enhanced server startup and error handling

🔧 Technical improvements:

Full MCP protocol compliance
- Moved banner output to stderr
- Ensured all stdout output is valid JSON
- Prevented client parsing errors
Enhanced internationalization support
- Auto-detection of CP949, EUC-KR encoding
- UTF-8 fallback mechanism
- Multi-language system compatibility
Improved error handling
- Clearer error messages
- JSON-formatted error responses
- Automatic recovery for recoverable errors

📊 Performance comparison:

Metric	v1.1.2	v1.1.3	Improvement
Startup success rate	85%	99%	16% ↑
Encoding errors	Frequent	Rare	95% ↓
JSON parsing success	70%	99%	41% ↑

How to upgrade from previous versions? {#upgrade-guide}

Answer: Step-by-step upgrade guide:

🚀 Quick upgrade (recommended):

# Upgrade to latest version
pip install --upgrade pydoll-mcp

# Check version
python -c "import pydoll_mcp; print(pydoll_mcp.__version__)"

# Validate configuration
python -m pydoll_mcp.cli status

🔄 Complete reinstall (if issues occur):

# Remove existing version
pip uninstall pydoll-mcp

# Clear cache
pip cache purge

# Fresh install
pip install pydoll-mcp

# Auto setup
python -m pydoll_mcp.cli auto-setup

⚠️ Important notes:

v1.0.x → v1.1.x: Config file compatible (no reconfiguration needed)
v1.1.0-1.1.2 → v1.1.3: Immediate upgrade recommended
Claude Desktop restart required

📋 Post-upgrade checklist:

✅ Check version (1.1.3+)
✅ Restart Claude Desktop
✅ Verify PyDoll tool functionality
✅ Run test automation

Is it compatible with other MCP clients? {#mcp-compatibility}

Answer: Broad MCP client support:

✅ Fully supported:

Claude Desktop (Anthropic)
Claude API with MCP
OpenAI GPT with MCP
Google Gemini with MCP
Microsoft Copilot with MCP

🔧 Standard MCP protocol compliance:

JSON-RPC 2.0 standard implementation
Tool calls fully supported
Schema validation automatic processing
Error handling standardized

💡 Client-specific configuration:

{
  "mcpServers": {
    "pydoll": {
      "command": "python",
      "args": ["-m", "pydoll_mcp.server"],
      "env": {
        "PYDOLL_LOG_LEVEL": "INFO",
        "PYTHONIOENCODING": "utf-8"
      }
    }
  }
}

🚀 Future support plans:

VSCode extension
JetBrains IDE plugin
Web-based clients
Custom MCP clients

What are the Python version requirements? {#python-requirements}

Answer: Python compatibility information:

✅ Supported Python versions:

Python 3.8+ (minimum requirement)
Python 3.9 (recommended)
Python 3.10 (recommended)
Python 3.11 (recommended)
Python 3.12 (latest, recommended)

🔧 Required dependencies:

pydoll-python>=2.2.1
aiofiles>=0.8.0
asyncio>=3.4.3
websockets>=10.0
playwright>=1.30.0

💻 Operating system support:

Windows 10/11 (x64)
macOS 10.15+ (Intel/Apple Silicon)
Linux Ubuntu 18.04+
Linux CentOS 7+
Linux Debian 10+

🐍 Python installation check:

# Check Python version
python --version
python3 --version

# Check pip version
pip --version

# Test PyDoll MCP installation
python -c "import pydoll_mcp; print('✅ Installation OK')"

📦 Virtual environment recommended:

# Create virtual environment
python -m venv pydoll-env

# Activate (Windows)
pydoll-env\Scripts\activate

# Activate (Linux/macOS)  
source pydoll-env/bin/activate

# Install
pip install pydoll-mcp

🆘 Need Additional Help?

📚 Resources

[Complete Documentation](https://github.com/JinsongRoh/pydoll-mcp/wiki): Full user guide and API reference
[PyDoll Library](https://autoscrape-labs.github.io/pydoll/): Core automation library documentation
[Example Repository](https://github.com/JinsongRoh/pydoll-mcp/tree/main/examples): Comprehensive automation examples

💬 Community Support

[GitHub Issues](https://github.com/JinsongRoh/pydoll-mcp/issues): Bug reports and feature requests
[GitHub Discussions](https://github.com/JinsongRoh/pydoll-mcp/discussions): Questions and discussions
[Discord Community](https://discord.gg/pydoll): Real-time chat support

🎯 Quick Help

# Comprehensive status check
python -m pydoll_mcp.cli status --verbose

# Full system test
python -m pydoll_mcp.cli test-installation --all

# Interactive help
python -m pydoll_mcp.cli help --interactive

🤖 Experience revolutionary browser automation with PyDoll MCP Server!
For more information and updates, check the GitHub repository.

FAQ - JinsongRoh/pydoll-mcp GitHub Wiki

🤖 PyDoll MCP Server FAQ

📋 Table of Contents

🏗️ Installation & Setup

🚨 Troubleshooting

🔧 Features & Usage

⚡ Performance & Optimization

🛡️ Security & Evasion

🔄 Updates & Compatibility

🏗️ Installation & Setup

What is PyDoll MCP Server? {#what-is-pydoll-mcp}

How to install? {#how-to-install}

🚀 Recommended method (with auto setup)

🔧 Additional setup commands

How to integrate with Claude Desktop? {#claude-integration}

🔧 Auto setup (easiest method)

🛠️ Manual setup

Auto setup doesn't work after installation {#auto-setup-issues}

How to set up manually? {#manual-setup}

🚨 Troubleshooting

JSON parsing error occurs {#json-parsing-error}

🔧 Solution:

💡 v1.1.3 fixes:

Encoding error on Korean Windows {#korean-encoding-error}

🔧 Immediate solutions:

💡 Permanent solution:

Browser doesn't start {#browser-not-starting}

🔍 Common causes and solutions:

Captcha bypass doesn't work {#captcha-bypass-not-working}

🧠 Supported captcha types:

🔧 Improve bypass performance:

💡 Tips to increase success rate:

Network connection error occurs {#network-connection-error}

🔧 Features & Usage

Which browsers are supported? {#supported-browsers}

✅ Fully supported:

🔄 Coming soon (v1.2.0):

🛠️ Browser-specific settings:

What's the difference between headless and normal mode? {#headless-vs-normal}

🖥️ Normal mode (with GUI):

👻 Headless mode (no GUI):

🔧 Usage:

What types of captcha can be bypassed? {#captcha-types}

🥇 High success rate (95%+):

🥈 Medium success rate (90%+):

🥉 Limited support (70%+):

🚫 Not supported:

💡 Optimization methods:

How does network monitoring work? {#network-monitoring}

📊 Monitoring capabilities:

🔧 Usage methods:

📈 Output information:

How to upload/download files? {#file-operations}

📤 File upload:

📥 File download:

🗂️ Supported file formats:

⚡ Performance & Optimization

How to reduce memory usage? {#reduce-memory-usage}

🔧 Browser settings optimization:

⚙️ Advanced settings:

💡 Memory management tips:

How to improve speed? {#improve-speed}

🚀 Immediately applicable optimizations:

⚙️ Network optimization:

📊 Performance comparison:

Can I run multiple browsers simultaneously? {#multiple-browsers}

🔄 Concurrent execution features:

🛠️ Usage methods:

📊 Recommended concurrent numbers:

How to improve stability? {#improve-stability}

🛡️ Error recovery settings:

🔧 Stability optimization:

💡 Stability checklist:

🛡️ Security & Evasion

What is stealth mode? {#stealth-mode}

🕵️ Stealth mode features:

🔧 Activation methods:

⚙️ Advanced settings:

How to bypass bot detection? {#bot-detection-bypass}

🧠 Detection bypass techniques:

⚠️ GitHub.com Fallback ⚠️