ReMind is an intelligent assistant that captures screen context and helps you manage tasks, generate summaries, and interact with AI based on what you're working on. It features a modern SvelteKit frontend with a FastAPI backend.
- Screen Context Recording: Capture and process screen content
- AI Chat Interface: Interactive chat with context awareness
- Task Management: Automatic task extraction and execution
- Smart Summaries: Generate summaries from your work context
- Prompt Suggestions: Context-aware prompt recommendations
- User Profiles: Personalized onboarding and settings
- Dark Mode: Full dark mode support
- Frontend: SvelteKit, TypeScript, Tailwind CSS
- Backend: FastAPI (Python), Uvicorn
- AI Integration: Support for OpenAI, Anthropic, Dify workflows
- Task Automation: Feishu/Lark integration ready
- Node.js 18+ and npm
- Python 3.9+
- Git
# Install dependencies
npm install
# Run development server
npm run devFrontend will be available at http://localhost:5173
# Navigate to backend directory
cd backend
# Install Python dependencies
pip install -r requirements.txt
# Run the server
python main.pyBackend API will be available at http://localhost:8000
# Make the script executable
chmod +x start.sh
# Run both frontend and backend
./start.sh-
Start the backend server:
cd backend && python main.py
-
In a new terminal, start the frontend:
npm run dev
-
Open http://localhost:5173 in your browser
Create a .env file in the backend directory:
# Optional: Dify API Integration
USE_DIFY_API=false
DIFY_API_KEY=your_api_key
DIFY_USER_ID=default-userThe app includes a development mode that clears localStorage on refresh. Toggle this in src/routes/+page.svelte:
const DEV_MODE = false; // Set to false for productionProject_Cortex/
โโโ src/ # Frontend source code
โ โโโ lib/
โ โ โโโ components/ # Svelte components
โ โ โโโ stores/ # Svelte stores
โ โ โโโ utils/ # Utility functions
โ โโโ routes/ # SvelteKit routes
โโโ backend/ # Backend API server
โ โโโ main.py # FastAPI application
โ โโโ dify_*.py # Dify integration modules
โ โโโ requirements.txt # Python dependencies
โโโ static/ # Static assets
โโโ package.json # Node.js dependencies
- ChatPanel: Main chat interface
- TaskPanel: Task management and execution
- SummaryPanel: Context summaries
- OnboardingModal: User onboarding flow
- ProfileSettings: User profile management
- RecordingIndicator: Screen recording status
/api/record/*- Screen recording control/api/context/*- Context processing/api/chat- AI chat interaction/api/tasks/*- Task management/api/summary/*- Summary generation/api/prompts/*- Prompt suggestions
- OpenAI GPT models
- Anthropic Claude
- Google Gemini
- Custom LLMs via Dify
- Feishu/Lark
- Email services
- Calendar applications
- Custom webhooks
# Build frontend
npm run build
# Preview production build
npm run previewnpm run check# Run type checking in watch mode
npm run check:watchMIT
Contributions are welcome! Please feel free to submit a Pull Request.
- Mock data is used by default (configure Dify for real integrations)
- Screen recording requires platform-specific implementation
For issues and questions, please use the GitHub issues tracker.