Friday Assistant is a next-generation, voice-activated Personal AI companion built with Flutter. Designed with a "Sci-Fi First" philosophy, it combines state-of-the-art Large Language Models (LLMs) with a premium, immersive user interface featuring holographic aesthetics, reactive animations, and proactive intelligence.
Unlike standard chatbots, Friday possesses Long-Term Memory, Real-Time Web Access, and a Proactive Sentinel Mode that watches out for you even when the app is in the background.
| Feature | Benefit |
|---|---|
| Voice-first interface | Hands-free operation |
| Long-term memory | Remembers your preferences across sessions |
| Proactive notifications | Alerts you before conflicts arise |
| Tool integration | Opens apps, sends messages, navigates |
| Multi-platform | Android, iOS, Web, Desktop |
| Privacy-focused | Your data stays in your Supabase vault |
-
The Friday Orb: A central, living core that reacts to voice amplitude, pulses when thinking, and transforms based on AI state (Idle, Listening, Processing, Speaking)
-
Holographic Aesthetics: Glassmorphism effects, neon gradients, and dynamic Particle Field backgrounds
-
Neural HUD: Real-time system status display with Neural Link indicator and active Memory Node count
-
Immersive Feedback: Haptic feedback and distinct sound effects for all interactions
-
FridayBrainService: Powered byGroq (Llama 3.3 70B) for lightning-fast inference and complex reasoning
-
Agentic Workflow: Visualized thought process phases: Thinking -> Searching Web -> Accessing Memory -> Synthesizing
-
Memory Auditor: Autonomous sub-agent that verifies information against existing memory to prevent conflicts
-
Tool Integration: - web_search: Real-time data fetching (Weather, News, Stock Prices)
- save_memory/search_memory: Supabase Vault integration for personal facts and preferences
-
Background Nudges: Scheduled background checks on schedule, weather, and environment
-
Contextual Awareness: Intelligent notifications (e.g., "Outdoor meeting at 3 PM with forecasted rain - take an umbrella")
-
Manual Diagnostics: Trigger instant Sentinel Check for immediate analysis
-
Multilingual Support: English, French, German, Spanish, and Hindi
-
Real-Time Transcription: Seamless speech-to-text dictation
-
Natural Voice: High-quality Text-to-Speech response system
-
Local Processing: Initial reasoning via Groq's high-speed inference
-
Encrypted Vault: Personal memories in private Supabase instance with Row Level Security (RLS)
-
Minimal Footprint: No audio data storage; STT processed and discarded in real-time
| Component | Technology |
|---|---|
| Framework | Flutter (Dart) |
| AI Backend | Groq API (Llama 3.3 70B) |
| Database | Supabase (PostgreSQL) |
| State Management | Native setState & Streams |
| Animations | flutter_animate & Custom Painters |
| Voice | speech_to_text & flutter_tts |
| Background Tasks | workmanager |
| Notifications | flutter_local_notifications |
- Flutter SDK v3.10+ installed
- Supabase Project with
chat_messagesandFriday_memorytables - Groq API Key for LLM inference
Clone the Repository
git clone https://github.com/dj2313/jarvis_assistant.git
cd jarvis_assistant
Configure Environment Create a.env file in the root directory:
GROQ_API_KEY=your_groq_api_key_here
SUPABASE_URL=your_supabase_url_here
SUPABASE_ANON_KEY=your_supabase_anon_key_here
Install Dependencies
flutter pub get
Run the App
flutter run
lib/
├── core/ # Configuration and Constants
├── models/ # Data Models (ChatMessage, etc.)
├── screens/ # UI Screens
│ ├── friday_home_screen.dart # Main HUD Interface
│ ├── friday_initialization_screen.dart # Boot-up Sequence
│ └── ...
├── services/ # Logic & API Layers
│ ├── friday_brain_service.dart # AI Core, Groq Integration, & Tools
│ ├── memory_service.dart # Supabase DB Interactions
│ └── vocal_service.dart # TTS & STT Handling
└── widgets/ # Reusable UI Components
└── friday_orb.dart # The animated AI Core
Contributions are welcome! See CONTRIBUTING.md for guidelines.
This project is licensed under the MIT License - see LICENSE for details.
If you find this project useful, please give it a star!
Built with love for the Future.