AI-Powered Personal Growth Companion
An intelligent mentor and coach for goal tracking, habit building, and personal development.
MentorMe is a comprehensive personal growth application that combines goal tracking, habit building, and journaling with AI-powered coaching. Choose between Claude's advanced cloud-based language models or privacy-focused local AI running entirely on your device. MentorMe provides personalized guidance, insights, and motivation to help you achieve your goals.
- Flexible AI Options: Choose between cloud-based Claude API or on-device local AI models
- Intelligent Mentoring: AI analyzes your progress and provides contextual coaching
- Goal Management: Track goals with milestones, deadlines, and progress monitoring
- Habit Building: Build lasting habits with streak tracking and smart reminders
- Reflective Journaling: Capture thoughts, moods, and energy levels with guided prompts
- Pulse Check-ins: Quick mood and energy tracking throughout the day
- Chat Interface: Natural conversations with your AI mentor
- Privacy-First: All data stored locally, with optional offline AI for complete privacy
- Create and organize goals with categories (Health, Career, Personal, etc.)
- Break down goals into actionable milestones
- Set deadlines and track progress
- Visual progress indicators
- Goal analytics and insights
- Daily habit tracking with streak monitoring
- Smart reminders to maintain consistency
- Habit analytics showing patterns and trends
- Celebration of milestones (7-day, 30-day streaks)
- Flexible habit scheduling
- Free-form and guided journaling options
- Mood and energy level tracking
- Link journal entries to goals
- Search and filter past entries
- AI-powered insights from your journal
- Personalized coaching cards based on your activity
- Context-aware advice and motivation
- Pattern detection (stalled goals, broken streaks, etc.)
- Natural language chat with AI mentor
- Goal decomposition and planning assistance
- Quick check-ins throughout the day
- Custom pulse types (mood, energy, stress, focus, etc.)
- Historical trends and visualizations
- Identify patterns affecting your progress
- Adaptive notification system
- Morning and evening check-in prompts
- Habit reminders at optimal times
- Streak protection alerts
- Deadline notifications
- Local-first data storage
- Export/import functionality for backups
- No cloud sync required (optional)
- Full control over your data
- Flutter - Cross-platform UI framework (web + mobile)
- Dart 3.0+ - Modern, null-safe programming language
- Provider - State management
- SharedPreferences - Local data persistence
MentorMe offers flexible AI options to suit your privacy and performance needs:
Remote AI (Claude API)
- Anthropic's advanced language models via cloud API
- Support for multiple models (Opus 4, Sonnet 4.5, Sonnet 4, Haiku 4)
- Requires internet connection and API key
- Custom proxy server for web CORS handling
Local AI (On-Device)
- Privacy-first: All processing happens on your device
- No internet required for AI features
- Google LiteRT LLM (AI Edge) for local model execution
- Gemma 3-1B-IT model optimized for on-device inference
- Downloadable models run entirely offline
- Requires Android 12+ with 4GB+ RAM (6GB+ recommended)
http- API communicationintl- Internationalization and date formattingandroid_alarm_manager_plus- Precise alarm schedulingflutter_local_notifications- Local notification systempath_provider&universal_io- File system accessshare_plus- Data sharing capabilities
Don't want to build from source? Download the latest APK directly:
- Android 12.0 or higher (check Settings → About Phone → Android version)
- ARMv8 64-bit processor
- 4GB+ RAM
Installation:
- Go to Releases
- Download
mentor-me-vX.X.X.apk - Install on your Android device (requires enabling "Install from Unknown Sources")
- Open the app and start your personal growth journey! 🎉
For detailed installation instructions and troubleshooting, see the APK Download Guide.
- Flutter SDK 3.0.0 or higher (Install Flutter)
- Dart SDK 3.0.0 or higher (included with Flutter)
- Node.js 14+ (for web proxy server - web development only)
- Claude API Key (optional - only needed for remote AI) (Get API Key)
Note: You can use MentorMe completely offline with local AI models (Android only), or use Claude's cloud API for more advanced capabilities.
Minimum Requirements:
- Android 12+ (API level 31 or higher) - Required by LiteRT LLM library
- 4GB RAM minimum (6GB+ strongly recommended)
- 2GB free storage for Gemma 3-1B-IT model
- ARMv8 64-bit processor
Recommended for Best Performance:
- Android 13+
- 6-8GB RAM
- Modern mid-range to flagship devices (e.g., Pixel 6+, Samsung Galaxy S21+, OnePlus 9+, or equivalent)
- GPU acceleration support for faster inference
Note: Local AI performance varies by device. Newer devices with more RAM and better processors will provide faster, more responsive AI interactions. The Gemma 3-1B-IT model is optimized for on-device use and works well on modern mid-range Android devices.
-
Clone the repository
git clone https://github.com/snowch/habits.git cd habits -
Install Flutter dependencies
flutter pub get
-
Set up the proxy server (for web development)
cd proxy npm install -
Configure AI Provider (Optional)
- For Remote AI: Enter your Claude API key on first launch or via Settings > AI Settings
- For Local AI: Skip the API key and select "Local AI" in Settings > AI Settings
- You can switch between remote and local AI anytime in settings
# Terminal 1: Start the proxy server (required for web)
cd proxy
npm start
# Terminal 2: Run Flutter web app
flutter run -d chromeThe proxy server runs at http://localhost:3000 and handles CORS for Claude API calls.
# Run on connected device or emulator
flutter run
# List available devices
flutter devicesNote: Mobile builds don't require the proxy server (direct API calls).
flutter build web
# Output: build/web/# Using the provided build script
./build_android.sh
# Or manually
flutter build apk --release
flutter build appbundle --releaselib/
├── main.dart # App entry point
├── models/ # Data models
│ ├── goal.dart
│ ├── habit.dart
│ ├── journal_entry.dart
│ ├── checkin.dart
│ ├── pulse_entry.dart
│ ├── pulse_type.dart
│ └── chat_message.dart
├── providers/ # State management
│ ├── goal_provider.dart
│ ├── habit_provider.dart
│ ├── journal_provider.dart
│ ├── checkin_provider.dart
│ ├── pulse_provider.dart
│ └── chat_provider.dart
├── screens/ # UI screens
│ ├── home_screen.dart
│ ├── goals_screen.dart
│ ├── habits_screen.dart
│ ├── journal_screen.dart
│ ├── mentor_screen.dart
│ ├── chat_screen.dart
│ └── settings_screen.dart
├── services/ # Business logic
│ ├── ai_service.dart
│ ├── storage_service.dart
│ ├── notification_service.dart
│ ├── mentor_intelligence_service.dart
│ └── backup_service.dart
├── widgets/ # Reusable widgets
├── theme/ # App theming
└── constants/ # Constants and strings
└── app_strings.dart
proxy/ # CORS proxy for web
├── server.js
└── package.json
docs/ # Documentation
└── MENTOR_LOGIC.md # AI mentor logic details
MentorMe supports both remote and local AI models. Configure via Settings > AI Settings:
Remote Models (Claude API)
- Claude Opus 4 - Most capable, best for complex coaching
- Claude Sonnet 4.5 - Balanced performance and capability
- Claude Sonnet 4 - Fast and efficient (default)
- Claude Haiku 4 - Quick responses, lower cost
Local Models (On-Device)
- Gemma 3-1B-IT - Google's instruction-tuned language model running via LiteRT
- Complete Offline Operation - No internet required after initial download
- Privacy-First - All data and processing stays on your device
- Optimized for Mobile - Efficient model designed for on-device performance
To use local AI:
- Navigate to Settings > AI Settings
- Select "Local AI" as your AI provider
- Download the Gemma 3-1B-IT model (~2GB, one-time download)
- Accept the HuggingFace license agreement
- Wait for download to complete
- Start using AI features completely offline
Technology: Powered by Google LiteRT LLM (AI Edge library) with the Gemma 3-1B-IT model from HuggingFace.
Device Requirements: Requires Android 12+ with at least 4GB RAM (6GB+ recommended). See Prerequisites for detailed requirements.
Configure check-in reminders and habit notifications:
- Navigate to Settings > Mentor Reminders
- Enable/disable morning and evening check-ins
- Set preferred notification times
- Configure habit-specific reminders
Export your data for safekeeping:
- Settings > Backup & Restore
- Export Data → Downloads JSON backup
- Import Data → Restore from backup file
# Run all tests
flutter test
# Run with coverage
flutter test --coverage
# Run specific test file
flutter test test/models/goal_test.dart# Analyze code
flutter analyze
# Format code
flutter format lib/
# Clean build artifacts
flutter cleanFor detailed development guidelines, architecture decisions, and best practices, see:
- CLAUDE.md - Complete developer guide
- docs/MENTOR_LOGIC.md - AI mentor intelligence system
- proxy/README.md - Proxy server documentation
- Multi-language support (i18n)
- Cloud sync (optional)
- Team/shared goals
- Advanced analytics dashboard
- Integration with calendar apps
- Voice journaling
- Goal templates library
Contributions are welcome! Please follow these steps:
- Fork the repository
- Create a feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
- Follow Flutter best practices outlined in CLAUDE.md
- Write tests for new features
- Update documentation as needed
- Use the
AppStringsclass for user-facing text - Ensure null safety compliance
- Run
flutter analyzebefore submitting
Problem: CORS errors when calling Claude API from web
Solution: Ensure the proxy server is running:
cd proxy
npm startProblem: Reminders not appearing on Android
Solution:
- Check Settings > Mentor Reminders - ensure enabled
- Grant notification permissions in device settings
- Disable battery optimization for MentorMe app
Problem: Build errors after git pull
Solution:
flutter clean
flutter pub get
flutter pub run build_runner build --delete-conflicting-outputsThis project is licensed under the GNU General Public License v3.0 - see the LICENSE file for details.
Copyright (C) 2025 Chris Snow
- Anthropic - Claude API for AI capabilities
- Flutter Team - Excellent cross-platform framework
- Open Source Community - All the amazing packages used in this project
Chris Snow
For bugs, feature requests, or questions, please open an issue on GitHub.
Built with ❤️ using Flutter and Claude AI









