Skip to content

README.md

Latest commit

 

History

History
174 lines (139 loc) · 4.22 KB

README.md

File metadata and controls

174 lines (139 loc) · 4.22 KB

ReMind - AI-Powered Screen Context Assistant

🚀 Overview

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.

✨ Features

  • 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

🛠 Tech Stack

  • Frontend: SvelteKit, TypeScript, Tailwind CSS
  • Backend: FastAPI (Python), Uvicorn
  • AI Integration: Support for OpenAI, Anthropic, Dify workflows
  • Task Automation: Feishu/Lark integration ready

📦 Installation

Prerequisites

  • Node.js 18+ and npm
  • Python 3.9+
  • Git

Frontend Setup

# Install dependencies
npm install

# Run development server
npm run dev

Frontend will be available at http://localhost:5173

Backend Setup

# Navigate to backend directory
cd backend

# Install Python dependencies
pip install -r requirements.txt

# Run the server
python main.py

Backend API will be available at http://localhost:8000

🚀 Quick Start

Using the Start Script

# Make the script executable
chmod +x start.sh

# Run both frontend and backend
./start.sh

Manual Start

  1. Start the backend server:

    cd backend && python main.py

  2. In a new terminal, start the frontend:

    npm run dev

  3. Open http://localhost:5173 in your browser

🔧 Configuration

Environment Variables

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-user

Development Mode

The 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 production

📁 Project Structure

Project_Cortex/
├── src/                    # Frontend source code
│   ├── lib/               
│   │   ├── components/    # Svelte components
│   │   ├── stores/        # Svelte stores
│   │   └── utils/         # Utility functions
│   └── routes/            # SvelteKit routes
├── backend/               # Backend API server
│   ├── main.py           # FastAPI application
│   ├── dify_*.py         # Dify integration modules
│   └── requirements.txt  # Python dependencies
├── static/                # Static assets
└── package.json          # Node.js dependencies

🎯 Key Components

Frontend Components

  • ChatPanel: Main chat interface
  • TaskPanel: Task management and execution
  • SummaryPanel: Context summaries
  • OnboardingModal: User onboarding flow
  • ProfileSettings: User profile management
  • RecordingIndicator: Screen recording status

Backend Endpoints

  • /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

🔌 Integration Points

AI Services

  • OpenAI GPT models
  • Anthropic Claude
  • Google Gemini
  • Custom LLMs via Dify

Task Automation

  • Feishu/Lark
  • Email services
  • Calendar applications
  • Custom webhooks

🧪 Development

Build for Production

# Build frontend
npm run build

# Preview production build
npm run preview

Type Checking

npm run check

Code Quality

# Run type checking in watch mode
npm run check:watch

📝 License

MIT

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

🐛 Known Issues

  • Mock data is used by default (configure Dify for real integrations)
  • Screen recording requires platform-specific implementation

📮 Support

For issues and questions, please use the GitHub issues tracker.