TagUI: Free RPA Tool by AI Singapore

Key Features:
- Free and open-source RPA tool
- Works on Windows, macOS, and Linux
- Web automation (Chrome, Edge)
- Desktop automation with visual recognition
- OCR capabilities for text extraction
- Supports 23 languages for writing flows
- Python, R, and JavaScript integration
- No setup needed in most environments

Core Technologies:
- Languages: JavaScript, PHP, Python
- Web Automation: Chrome DevTools Protocol
- Visual Automation: SikuliX
- OCR: Tesseract OCR
- Legacy: CasperJS, PhantomJS, SlimerJS
- Package Managers: NPM, Python pip
- Browser Support: Chrome, Edge (native)

# Check Java version (required for visual automation)
java -version  # Java JDK 8+ (64-bit)

# Check Chrome is installed
google-chrome --version  # or 'chrome --version' on Windows

# Optional: Check Python version (for Python package)
python --version  # Python 3.6+

# Optional: Check Node.js (for NPM package)
node --version  # Node.js 12+

# Download from GitHub Releases
# Visit: https://github.com/aisingapore/TagUI/releases

# For Windows: Download TagUI_Windows.zip
# For macOS: Download TagUI_macOS.zip
# For Linux: Download TagUI_Linux.tar.gz

# Extract to a folder without spaces in the path
# Windows: Extract to C:\tagui
# macOS: Extract to ~/Desktop/tagui
# Linux: Extract to /home/your_userid/tagui

# 1. Download and install OpenJDK 8+ (64-bit)
# Visit: https://adoptium.net/ and download the latest JDK

# 2. Extract TagUI to C:\tagui (avoid spaces in path)

# 3. Add C:\tagui\src to your system PATH
# - Right-click 'This PC' → Properties → Advanced system settings
# - Click 'Environment Variables'
# - Under 'System variables', find 'Path' and click Edit
# - Click 'New' and add: C:\tagui\src
# - Click OK to save

# 4. Open a new Command Prompt and test
tagui samples\1_google

# 1. Download TagUI_macOS.zip from GitHub Releases

# 2. Extract to Desktop
cd ~/Desktop
unzip TagUI_macOS.zip

# 3. Add TagUI to PATH (add to ~/.zshrc or ~/.bash_profile)
echo 'export PATH="$HOME/Desktop/tagui/src:$PATH"' >> ~/.zshrc
source ~/.zshrc

# 4. Install Java JDK 8+ if needed (for visual automation)
# Download from: https://adoptium.net/

# 5. Test TagUI
tagui samples/1_google

# 1. Download TagUI_Linux.tar.gz from GitHub Releases

# 2. Extract to home directory
cd ~
tar -xzf TagUI_Linux.tar.gz

# 3. Install OpenJDK (Ubuntu/Debian)
sudo apt update
sudo apt install openjdk-11-jdk

# Or for Fedora/RHEL
sudo dnf install java-11-openjdk-devel

# 4. Add TagUI to PATH (add to ~/.bashrc)
echo 'export PATH="$HOME/tagui/src:$PATH"' >> ~/.bashrc
source ~/.bashrc

# 5. Make TagUI executable
chmod +x ~/tagui/src/tagui

# 6. Test TagUI
tagui samples/1_google

# Install TagUI Python package
pip install tagui

# Or install RPA for Python (optimized fork)
pip install rpa

# Verify installation
python -c "import tagui as t; print('TagUI installed successfully')"

# Or for RPA package
python -c "import rpa as r; print('RPA installed successfully')"

# Install TagUI globally via NPM
npm install -g tagui

# Or install locally in a project
npm install tagui

# Verify installation
tagui -v

# Run a sample flow
tagui samples/1_google

// Save this as google_search.tag

// Navigate to Google
https://www.google.com

// Type search query and press Enter
type search as automation testing[enter]

// Take a screenshot
snap page to results.png

// Click on the first result
click //h3[1]

// Wait 2 seconds
wait 2

# Run flow in visible mode (see browser)
tagui google_search.tag

# Run flow in headless mode (background)
tagui google_search.tag headless

# Run with Chrome browser (default)
tagui google_search.tag chrome

# Run with Edge browser
tagui google_search.tag edge

# Run with debugging output
tagui google_search.tag debug

# Run and generate report
tagui google_search.tag report

# Import TagUI
import tagui as t

# Initialize TagUI
t.init()

# Navigate to Google
t.url('https://www.google.com')

# Type search query and press enter
t.type('//*[@name="q"]', 'automation testing[enter]')

# Wait for results
t.wait(2)

# Take screenshot
t.snap('page', 'results.png')

# Read results count
results = t.read('resultStats')
print(f'Results: {results}')

# Close browser
t.close()

# Import TagUI
import tagui as t

# Initialize with visual automation enabled
t.init(visual_automation=True)

# Double-click on an icon (using image file)
t.dclick('outlook_icon.png')

# Wait for window to open
t.wait(3)

# Click the 'New Mail' button
t.click('new_mail_button.png')

# Type into a field
t.type('recipient_field.png', 'user@example.com[tab]')
t.type('subject_field.png', 'Automated Email[tab]')
t.type('body_field.png', 'This email was sent via TagUI automation.')

# Click send button
t.click('send_button.png')

# Close TagUI
t.close()

// Navigation
http://website.com or https://website.com
click element_identifier
rclick element_identifier  // right-click
dclick element_identifier  // double-click
hover element_identifier

// Input
type element_identifier as text to enter
select element_identifier as option value
upload element_identifier as file path

// Reading
read element_identifier to variable_name

// Verification
check element_identifier
uncheck element_identifier

// Screenshots
snap element_identifier to filename.png
snap page to filename.png

// Control Flow
wait 5  // wait 5 seconds
wait for element_identifier
if condition
  // commands
else
  // commands

// XPath selectors
click //button[@id='submit']
type //input[@name='username'] as john.doe

// CSS selectors
click button#submit
type input[name='username'] as john.doe

// HTML attributes
click submit  // matches id='submit'
type username as john.doe  // matches name='username'

// Text content
click 'Sign In'  // clicks element containing this text

// Image-based (visual automation)
click image_file.png

// Multiple selectors (first match wins)
click //button[@id='btn1'] | //button[@class='submit-btn']

# Example desktop automation flow (save as desktop_app.tag)

# Enable visual automation
visual on

# Double-click application icon
dclick app_icon.png

# Wait for app to load
wait 3

# Click menu item
click file_menu.png
click new_document.png

# Type text
type document_area.png as Hello World

# Save using keyboard shortcut
keyboard [ctrl]s
wait 1
type filename_field.png as mydocument
keyboard [enter]

# Take screenshot
snap page to completed.png

// Enable visual automation (includes OCR)
visual on

// Read text from a specific area
read invoice_total.png to total_amount
echo Total: `total_amount`

// Read text from entire screen area
read page.png to page_text

// Read from PDF (convert to image first)
read document.pdf to document_text

// English (default)
https://www.google.com
type search as hello world
click search button

// Chinese (中文)
网址 https://www.google.com
输入 搜索 为 你好世界
点击 搜索按钮

// Spanish (Español)
navegar https://www.google.com
escribir búsqueda como hola mundo
clic botón de búsqueda

// French (Français)
aller https://www.google.com
taper recherche comme bonjour monde
cliquer bouton rechercher

// Read CSV file and iterate through rows
table accounts.csv

// Loop through each row
for row in accounts
  https://example.com/login
  type username as `row.username`
  type password as `row.password`
  click login
  
  // Do work...
  
  click logout
  wait 2

// Embed JavaScript in TagUI flows

// Assign variables
current_time = new Date().toLocaleString()
echo Current time: `current_time`

// JavaScript code block
js begin
  var result = 0;
  for (var i = 1; i <= 10; i++) {
    result += i;
  }
  return result;
js finish

echo Sum of 1-10: `js_result`

// Call JavaScript functions
js console.log('Debug message')

# Run with debug output
tagui myflow.tag debug

# Check if element exists before interacting
if present('login_button')
  click login_button
else
  echo Login button not found

# Try-catch pattern using conditions
if exist('error_message.png')
  echo Error detected, taking screenshot
  snap page to error.png
  fail  // explicitly fail the flow

# View detailed logs
cat tagui_report.txt

# Install VS Code extension
# 1. Open VS Code
# 2. Go to Extensions (Ctrl+Shift+X)
# 3. Search for 'TagUI'
# 4. Click Install

# Or install via command line
code --install-extension tebelorg.tagui-language

# Features:
# - Syntax highlighting for .tag files
# - Code snippets for common commands
# - IntelliSense autocomplete
# - Error detection and warnings

// 1. Use descriptive variable names
user_email = 'john.doe@example.com'
login_timeout = 10

// 2. Add comments to explain complex logic
// Navigate to admin dashboard after login
https://app.example.com/admin

// 3. Use wait statements to handle timing issues
wait for dashboard_loaded

// 4. Take screenshots at critical steps
snap page to step1_login.png

// 5. Validate state before proceeding
if not present('logged_in_indicator')
  echo Login failed, aborting
  fail

// 6. Handle errors gracefully
if exist('error_notification')
  read error_notification to error_msg
  echo Error: `error_msg`
  snap page to error_state.png

// 7. Clean up resources
// Close all browser tabs at end
close

# Windows Task Scheduler
# 1. Open Task Scheduler
# 2. Create Basic Task
# 3. Set trigger (daily, weekly, etc.)
# 4. Action: Start a program
# 5. Program: C:\tagui\src\tagui.cmd
# 6. Arguments: C:\automation\myflow.tag headless

# Linux/macOS cron
# Edit crontab
crontab -e

# Run flow every day at 9 AM
0 9 * * * /home/user/tagui/src/tagui /home/user/flows/daily_report.tag headless

# Run flow every Monday at 8 AM
0 8 * * 1 ~/tagui/src/tagui ~/flows/weekly_backup.tag headless

# View cron logs (Linux)
grep CRON /var/log/syslog

# Issue: 'tagui' command not found
# Solution: Add TagUI to PATH or use full path
/path/to/tagui/src/tagui myflow.tag

# Issue: Visual automation not working
# Solution: Verify Java JDK is installed (64-bit)
java -version  # Should show 1.8 or higher

# Issue: Chrome/Edge not launching
# Solution: Update browser to latest version
google-chrome --version

# Issue: Element not found errors
# Solution: Add wait statements or use debug mode
tagui myflow.tag debug

# Issue: Permissions error on Linux/macOS
# Solution: Make tagui executable
chmod +x ~/tagui/src/tagui

# Clean cache and retry (Python package)
pip uninstall tagui -y
pip install tagui --no-cache-dir

# View detailed error logs
cat tagui_report.txt

Official Resources:
- GitHub Repository: https://github.com/aisingapore/TagUI
- Documentation: https://tagui.readthedocs.io/
- Python Package (RPA): https://github.com/tebelorg/RPA-Python
- NPM Package: https://www.npmjs.com/package/tagui

Community:
- GitHub Issues: Report bugs and request features
- GitHub Discussions: Ask questions and share flows
- Stack Overflow: Tag questions with 'tagui'

Learning:
- Sample Flows: Check the samples/ folder in TagUI
- Video Tutorials: Search 'TagUI RPA tutorial' on YouTube
- AI Singapore Learn: https://learn.aisingapore.org/

Alternatives:
- RPA for Python: https://github.com/tebelorg/RPA-Python
- Selenium: For pure web automation
- Playwright: Modern browser automation