🎯 Automation Assassin

I was switching apps 10,888 times per week. This tool helped me save 3 hours/day.

Automation Assassin analyzes your macOS Screen Time data to detect productivity-killing patterns and generates personalized interventions that actually work.

🚀 Quick Start (5 minutes)

# 1. Clone and setup
git clone https://github.com/yourusername/automation-assassin
cd automation-assassin
pip install -r requirements.txt

# 2. Copy your Screen Time data
./copy_my_data.sh

# 3. Analyze your patterns
python analyze_my_patterns.py

# 4. Generate personalized interventions
python generate_my_interventions.py

# 5. Install interventions (requires Hammerspoon)
./install_interventions.sh

That's it! Your personalized productivity interventions are now active.

📊 What You'll Discover

When you run the analysis, you'll see your shocking statistics:

😱 Holy s***! You switched apps 10,888 times in 7 days!
⚡ 68.2% were bounce-backs (accidents)
💀 Your #1 death loop: Cursor → Safari (4,136 times)
⏰ Time lost to context switching: 21.3 hours/week
💰 That's worth $1,065/week at $50/hour

🛡️ What Gets Fixed

1. Bounce-Back Killer (68% of switches eliminated)

Detects and prevents accidental app switches when you overshoot with Cmd+Tab.

2. Smart Window Layouts (Your top 3 apps, always visible)

One hotkey arranges your most-used apps optimally, eliminating the need to switch.

3. AI Wait Optimizer (Productive waiting)

Detects when AI is processing and allows guilt-free app switching during wait times.

4. Real-time Statistics (Track your improvement)

Shows your productivity improvements compared to baseline.

📈 Real Results

From actual users:

Metric	Before	After	Improvement
Daily app switches	1,555	400	74% reduction
Bounce-backs	1,061/day	~0	100% eliminated
Time lost	3 hours/day	0.5 hours	2.5 hours saved
Deep work sessions	15 min	45 min	3x longer

🔧 How It Works

1. Analyzes YOUR Data: Reads your actual Screen Time database (stays local)
2. Identifies YOUR Patterns: Finds your specific death loops and time wasters
3. Generates YOUR Interventions: Creates custom Hammerspoon scripts for your workflow
4. Measures YOUR Progress: Tracks improvements against your baseline

📁 What Gets Generated

After running the analysis and generation scripts:

my_analysis/
├── report.md           # Your detailed productivity report
└── patterns.json       # Your app switching patterns

my_interventions/
├── complete_setup.lua  # All interventions combined
├── bounce_killer.lua   # Bounce-back prevention
├── smart_layouts.lua   # Window management
├── ai_detection.lua    # AI wait detection
└── stats_tracker.lua   # Statistics tracking

⌨️ Your Personal Hotkeys

After installation, you'll have these productivity superpowers:

• Cmd+Shift+S : Smart split-screen for your top 3 apps
• Cmd+Shift+F : Focused layout (70% main app)
• Cmd+Shift+G : AI processing mode (switch freely)
• Cmd+Shift+B : Toggle bounce protection
• Cmd+Shift+D : Show your productivity stats
• Cmd+Shift+I : Show AI wait statistics

🎬 See It In Action

Running the Analysis

$ python analyze_my_patterns.py

🔍 Analyzing your app usage patterns...

😱 Holy s***! You switched apps 10,888 times in 7 days!
⚡ 68.2% were bounce-backs (accidents)
💀 Your #1 death loop: Cursor → Safari (4136 times)
⏰ Time lost to context switching: 21.3 hours/week
💰 That's worth $1065/week at $50/hour

✅ Full report saved to: my_analysis/report.md
🚀 Next step: Run 'python generate_my_interventions.py'

Generating Interventions

$ python generate_my_interventions.py

🔧 Generating your personalized interventions...

📊 Loaded your analysis:
   • Total switches: 10,888
   • Bounce rate: 68.2%
   • Top apps: Cursor, Safari, Telegram

✅ Generated: my_interventions/complete_setup.lua
✅ Generated: my_interventions/bounce_killer.lua
✅ Generated: my_interventions/smart_layouts.lua

Expected Impact:
• Daily time saved: 2.5 hours
• Weekly switches eliminated: 7,426
• Estimated yearly value: $45,625

🧪 Testing With Sample Data

Don't have macOS or want to test first? Use the included test data:

# Use test fixtures instead of real data
cp tests/fixtures/screentime_test.db user_data/screentime.db
python analyze_my_patterns.py

🤖 Advanced: AI-Powered Analysis

For deeper insights using Claude AI (optional):

# Install Claude SDK
pip install claude-code-sdk

# Set API key
export ANTHROPIC_API_KEY=your-api-key

# Run AI-powered analysis
python src/run_analysis.py --ai-insights

📚 Documentation

• QUICKSTART.md - Detailed setup guide
• Architecture - How the system works
• Interventions - Available intervention types
• API Documentation - For developers

🛠️ Requirements

• macOS (for Screen Time data access)
• Python 3.9+
• Hammerspoon (free, for interventions)
• 5 minutes of your time

🔒 Privacy

• ✅ Your data stays local - nothing uploaded
• ✅ No tracking or analytics
• ✅ Open source - audit the code yourself
• ✅ You control everything

🤝 Contributing

Found a new pattern? Created a better intervention? PRs welcome!

Areas for improvement:

• Windows/Linux support
• More intervention types
• Browser extension integration
• Team productivity features

📖 The Full Story

Read about how I discovered I was wasting 3 hours/day and what I did about it: How I Saved 3 Hours/Day by Killing My 10,888 App Switches

⚠️ Disclaimer

This tool shows you the harsh reality of your app switching habits. The numbers might shock you. That's the point. Once you see the problem, you can fix it.

🙏 Acknowledgments

Built with:

• Hammerspoon - macOS automation
• Claude Code SDK - AI analysis (optional)
• macOS Screen Time - Usage data
• Lots of coffee and frustration with my own productivity

---

Stop switching. Start shipping. 🚀