Create beautiful websites with AI through simple chat interface

Hugo AI Studio is a modern web application that lets users create complete websites by simply describing what they want in natural language. The AI analyzes the description and generates a fully functional Hugo website with custom content.

• 💬 Simple Chat Interface - Just describe what website you want
• 🤖 AI-Powered Creation - Uses Ollama LLM to understand and create
• 🌐 Instant Preview - See your website immediately
• 📥 Download Ready - Get complete website files as ZIP
• 💾 Database Storage - All websites stored persistently
• 🐳 Docker Containerized - Easy deployment and scaling

┌─────────────────┐    ┌─────────────────┐    ┌─────────────────┐
│   React Chat    │    │  FastAPI + AI   │    │   Ollama LLM    │
│   Port: 3001    │◄──►│   Port: 8000    │◄──►│  Port: 11434    │
└─────────────────┘    └─────────────────┘    └─────────────────┘
         │                       │                       │
         └───────────────────────┼───────────────────────┘
                                 ▼
         ┌─────────────────┐    ┌─────────────────┐
         │  SQLite DB      │    │     Nginx       │
         │  (Persistent)   │    │   Port: 8080    │
         └─────────────────┘    └─────────────────┘

• Docker and Docker Compose
• 4GB+ RAM (for Ollama LLM)
• Internet connection (for initial setup)

git clone https://github.com/your-username/hugo-ai-studio.git
cd hugo-ai-studio

docker-compose -f compose.yml up -d

docker exec hugo-ai-studio-ollama-1 ollama pull llama3.2

• Chat Interface: http://localhost:3001
• API Documentation: http://localhost:8000/docs
• Generated Sites: http://localhost:8080/sites/{site-id}/

1. Open the chat interface at http://localhost:3001
2. Describe your website in natural language:
   • "Create a tech blog about artificial intelligence"
   • "Build a portfolio website for a photographer"
   • "Make a business website for a coffee shop"
3. Wait for AI to create your website (30-60 seconds)
4. Preview your site using the provided link
5. Download ZIP file with all website files

hugo-ai-studio/
├── frontend/                 # React Chat Interface
│   ├── src/
│   │   ├── App.js           # Main chat component
│   │   └── index.js         # React entry point
│   ├── public/
│   │   └── index.html       # HTML template
│   ├── package.json         # Dependencies
│   └── Dockerfile           # Frontend container
├── backend/                  # FastAPI Backend
│   ├── main.py              # API endpoints & AI logic
│   ├── requirements.txt     # Python dependencies
│   └── Dockerfile           # Backend container
├── compose.yml              # Docker Compose configuration
├── nginx.conf               # Nginx configuration
├── data/                    # SQLite database storage
├── generated-sites/         # Generated website files
└── ollama-data/            # LLM model storage

• REACT_APP_API_URL: Backend API URL (default: http://localhost:8000)
• OLLAMA_URL: Ollama service URL (default: http://ollama:11434)
• DATABASE_PATH: SQLite database path (default: /app/data/sites.db)

• 3001: React Frontend (Chat Interface)
• 8000: FastAPI Backend (API)
• 8080: Nginx (Generated Sites)
• 11434: Ollama (LLM Service)

# Frontend
cd frontend
npm install
npm start

# Backend
cd backend
pip install -r requirements.txt
uvicorn main:app --reload

# Ollama (separate terminal)
ollama serve
ollama pull llama3.2

• GET /health - Health check
• POST /api/create-website - Create website from description
• GET /api/download/{site_id} - Download website ZIP

"Create a tech blog about artificial intelligence and machine learning with posts about recent developments"

"Build a portfolio website for a graphic designer showcasing creative work and client testimonials"

"Make a professional website for a local coffee shop with menu, location, and contact information"

# Check service status
docker-compose -f compose.yml ps

# View logs
docker-compose -f compose.yml logs frontend
docker-compose -f compose.yml logs backend

# Verify model is downloaded
docker exec hugo-ai-studio-ollama-1 ollama list

# Re-download model
docker exec hugo-ai-studio-ollama-1 ollama pull llama3.2

# Stop all services
docker-compose -f compose.yml down

# Check port usage
netstat -tulpn | grep :3001

• CPU: 2+ cores recommended
• RAM: 4GB minimum, 8GB recommended
• Storage: 10GB for models and generated sites
• Network: Internet connection for initial setup

1. Fork the repository
2. Create feature branch (git checkout -b feature/amazing-feature)
3. Commit changes (git commit -m 'Add amazing feature')
4. Push to branch (git push origin feature/amazing-feature)
5. Open Pull Request

This project is licensed under the MIT License - see the LICENSE file for details.

• Hugo - Static site generator
• Ollama - Local LLM inference
• React - Frontend framework
• FastAPI - Backend framework
• Docker - Containerization

Made with ❤️ for easy website creation