Your Semantic Digital Twin - An intelligent system that captures, understands, and learns from your digital life.
ATLES-ECHO is a semantic monitoring system that creates a searchable, AI-powered memory of everything you do:
- π File Monitoring - Tracks code changes, documents, and project files
- π₯οΈ Screen Capture - OCR-based screen content extraction
- β¨οΈ Keystroke Monitoring - Context-aware typing patterns (privacy-focused)
- π Clipboard Tracking - Saves copied text and code snippets
- πͺ Application Monitoring - Tracks app usage and focus time
- π Semantic Search - Natural language search across all captured data
- π Auto-Generated Reports - Daily summaries, pattern analysis, productivity insights
- Backend: FastAPI, Python 3.11+
- Frontend: React 18, TypeScript, Vite, TailwindCSS
- Embeddings:
spartan8806/atles-champion-embedding(768-dim) - Vector DB: FAISS (Facebook AI Similarity Search)
- Knowledge Store: SQLite
- Monitoring: watchdog, mss, pytesseract, pynput, pywin32
- Python 3.11 or higher
- Node.js 18 or higher
- Tesseract OCR (for screen text extraction)
# Clone the repository
git clone https://github.com/spartan8806/atles-echo.git
cd atles-echo
# Install backend dependencies
cd backend
pip install -r requirements.txt
# Install frontend dependencies
cd ../frontend
npm installCreate a backend/.env file:
# API Configuration
API_PORT=5001
API_HOST=0.0.0.0
# Data paths
DATA_DIR=../data
WATCH_PATHS=["D:\\.atles"]
# Model configuration
EMBEDDING_MODEL=spartan8806/atles-champion-embedding
EMBEDDING_DIMENSION=768
# Feature toggles
ENABLE_SCREEN_MONITOR=true
ENABLE_KEYSTROKE_MONITOR=true
ENABLE_CLIPBOARD_MONITOR=true
ENABLE_APP_MONITOR=trueOption 1: Development Mode
# Terminal 1: Start backend
cd backend
python -m uvicorn main:app --reload --host 0.0.0.0 --port 5001
# Terminal 2: Start frontend
cd frontend
npm run devOption 2: Using the startup script
# Windows
.\start_echo.bat
# Linux/Mac
chmod +x start_echo.sh
./start_echo.shAccess the dashboard at: http://localhost:3000
- Installation Guide
- Configuration
- API Documentation
- Architecture Overview
- Privacy Analysis
- Cloudflare Tunnel Setup
IMPORTANT: ATLES-ECHO is designed for personal use only. All data stays on your machine.
- β
Local-first: All data stored locally in
./data/ - β No cloud uploads: Nothing leaves your machine unless you configure it
- β Full control: Delete data anytime, configure what's monitored
- β Open source: Audit the code yourself
By default, ATLES-ECHO monitors:
- File changes in configured directories
- Screen content (via OCR)
- Clipboard text (optional)
- Application usage (app names and window titles)
- Keystroke buffers (configurable whitelist)
You can disable any monitor via backend/config.py.
atles-echo/
βββ backend/ # FastAPI backend
β βββ main.py # Main API server
β βββ config.py # Configuration
β βββ models.py # Pydantic models
β βββ auth.py # Authentication
βββ core/ # Core functionality
β βββ embedding_engine.py
β βββ vector_db.py
β βββ knowledge_base.py
βββ watchers/ # Monitoring modules
β βββ file_watcher.py
β βββ screen_monitor.py
β βββ clipboard_monitor.py
β βββ keystroke_monitor.py
β βββ app_monitor.py
βββ frontend/ # React frontend
β βββ src/
β βββ pages/
β βββ components/
βββ docs/ # Documentation
Contributions are welcome! Please:
- 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
- Developers: Track code changes, debug sessions, research snippets
- Writers: Monitor document edits, research notes, writing patterns
- Researchers: Organize papers, notes, and reading sessions
- Knowledge Workers: Build a searchable memory of your work
- Multi-language support
- Browser history integration
- Email integration (local)
- Voice memo capture
- Advanced pattern detection
- LLM integration for insights
- Cross-platform support (macOS, Linux)
- Mobile companion app
MIT License - see LICENSE for details.
- Embedding Model: spartan8806/atles-champion-embedding
- FAISS: Facebook AI Research
- FastAPI: SebastiΓ‘n RamΓrez
- React: Meta
- π Issues: GitHub Issues
Made with β€οΈ for knowledge workers
"Your digital life, remembered."