MoodBoard

An AI-powered visual moodboard application. Drop images onto the canvas and get automatic AI-generated captions using Google Gemini.

Features

• Drag & Drop - Simply drop images from your computer onto the canvas
• AI Captioning - Images are automatically captioned using Google Gemini
• Visual Canvas - Arrange, drag, and organize your images freely
• Persistent Storage - Images and captions stored in MongoDB

Quick Start with Docker

1. Set your Gemini API key:

   export GEMINI_API_KEY=your_api_key_here

2. Start all services:

   make up

3. Open the app:

   • Frontend: http://localhost:3000
   • Backend API: http://localhost:8001

Development Setup

Prerequisites

• Python 3.13+
• Node.js 20+
• MongoDB (local or Docker)
• uv for Python package management

Install Dependencies

make install

Start MongoDB (if not running)

make mongo

Run Development Servers

export GEMINI_API_KEY=your_api_key_here
make dev

This starts:

• Backend at http://localhost:8001
• Frontend at http://localhost:5173

Available Commands

Run make help to see all available commands:

Command	Description
make install	Install all dependencies
make dev	Run both frontend and backend in dev mode
make dev-backend	Run backend only
make dev-frontend	Run frontend only
make build	Build Docker images
make up	Start all services with Docker
make down	Stop all services
make logs	View logs from all services
make clean	Remove build artifacts and volumes

Architecture

moodboard/
├── backend/
│   ├── main.py          # FastAPI app and routes
│   ├── database.py      # MongoDB operations
│   ├── captioning.py    # Gemini integration
│   └── models.py        # Pydantic schemas
├── frontend/
│   └── src/
│       └── App.tsx      # React application
├── docker-compose.yml   # Docker orchestration
└── Makefile            # Build automation

API Endpoints

Method	Endpoint	Description
POST	/upload	Upload an image for captioning
GET	/images	List all images with captions
GET	/images/{id}	Get a single image
DELETE	/images/{id}	Delete an image
GET	/health	Health check

Environment Variables

Variable	Description	Default
GEMINI_API_KEY	Google Gemini API key	Required
MONGODB_URL	MongoDB connection string	mongodb://localhost:27017

License

MIT