init-ai-app

Create AI-powered fullstack applications with Gemini integration.

Quick Start

npx init-ai-app@latest my-app
cd my-app
npm run install:all
# Update backend/.env with your GEMINI_API_KEY
npm run dev

Usage

Interactive Mode

Run the CLI without options to be guided through the setup:

npx init-ai-app@latest [project-name]

You'll be prompted to select:

1. Project name - Name of your project (default: my-app)
2. Backend - Node.js + Express (TypeScript)
3. Frontend - Angular or React
4. App type - Simple chat bot or Chat bot with file upload

Non-Interactive Mode

Skip prompts by providing options directly:

npx init-ai-app@latest my-app --frontend angular --backend node-express --app-type simple-chat

Hybrid Mode

Specify some options and get prompted for the rest:

npx init-ai-app@latest my-app --frontend react
# You'll be prompted for backend and app-type

Options

--frontend <framework>  Frontend framework (angular | react)
--backend <framework>   Backend framework (node-express)
--app-type <type>       Application type (simple-chat | chat-with-file-upload)
--use-npm               Use npm as package manager
--use-pnpm              Use pnpm as package manager
--use-yarn              Use yarn as package manager
--use-bun               Use bun as package manager
-V, --version           Output version number
-h, --help              Display help

Examples

# Full interactive (default)
npx init-ai-app@latest

# Angular + simple chat
npx init-ai-app@latest my-app --frontend angular --backend node-express --app-type simple-chat

# React + file upload with pnpm
npx init-ai-app@latest my-app --frontend react --backend node-express --app-type chat-with-file-upload --use-pnpm

# Just specify frontend, prompt for the rest
npx init-ai-app@latest my-app --frontend angular

Generated Project Structure

my-app/
├── backend/              # Node.js + Express API
│   ├── src/
│   │   ├── config/       # Gemini AI configuration
│   │   ├── controllers/  # Request handlers
│   │   ├── services/     # Business logic
│   │   ├── routes/       # API routes
│   │   └── types/        # TypeScript types
│   ├── .env              # Environment variables
│   └── package.json
├── frontend/             # Angular or React app
│   └── src/
├── .gitignore
├── README.md
└── package.json

API Endpoints

| Method | Endpoint | Description | |--------|----------|-------------| | GET | /api/health | Health check | | POST | /api/chat | Send message, get AI response | | POST | /api/chat/stream | Streaming response (SSE) | | POST | /api/files/upload | Upload file (file-upload variant) |

Environment Variables

Create a backend/.env file:

GEMINI_API_KEY=your-api-key-here
GEMINI_MODEL=gemini-3-flash-preview
PORT=3000

Get your API key from Google AI Studio.

Development

Prerequisites

• Node.js 18+
• npm, pnpm, yarn, or bun

Running the Generated App

# Install dependencies
npm run install:all

# Start development servers
npm run dev

• Backend: http://localhost:3000
• Frontend: http://localhost:4200 (Angular) or http://localhost:5173 (React)

Contributing

Local Development

# Clone the repository
git clone https://github.com/luixaviles/init-ai-app-boilerplate.git
cd init-ai-app

# Install dependencies
npm install

# Build the CLI
npm run build:ncc

# Test locally (interactive)
node dist/index.js test-app

# Test locally (non-interactive)
node dist/index.js test-app --frontend angular --backend node-express --app-type simple-chat

# Or link globally
npm link
init-ai-app test-app

Scripts

| Script | Description | |--------|-------------| | npm run build:ncc | Build CLI with ncc bundler | | npm run clean | Remove dist folder |

Tech Stack

CLI

• Commander - CLI framework
• Prompts - Interactive prompts
• fast-glob - File pattern matching
• fs-extra - File system utilities
• @vercel/ncc - Zero-dependency bundling

Generated Backend

• Node.js + Express
• TypeScript
• @google/genai - Gemini AI SDK

Generated Frontend

• Angular 21 (standalone components) or React 19 + Vite
• TypeScript

License

MIT