Mini App Factory

A powerful AI-powered web application that generates production-ready multi-page websites from natural language descriptions. Simply describe what you want to build, and the AI will create a complete, responsive project for you.

Features

• Natural Language Input: Describe your website in plain English
• Multi-page Architecture: Automatically generates multiple linked pages based on description
• Two-Pass AI Pipeline:
  • Conceptualize: AI creates a detailed design specification and site map
  • Generate: AI builds production-ready files with Tailwind CSS v4
• Project Dashboard: Manage and track all your generated sites (including redeploys)
• Authentication: Secure access to your projects via Stack Auth
• AI Control Panel: Per-user persisted AI settings (BYOK + model/provider controls)
• Admin-Only Controls: Global provider controls restricted to aditya.mer@somaiya.edu
• Live Preview & Editor: Interactive workspace to see and edit your sites in real-time
• Download & Deploy: Export as ZIP with all assets or deploy to the web
• Modern UI: Built with Tailwind CSS v4 and shadcn/ui components

How It Works

1. Input Your Description: Tell the AI what kind of website you want to create.
2. AI Conceptualizes: The AI analyzes your request, creates a design spec, and plans the page structure.
3. AI Generates: The AI generates all necessary HTML, CSS (Tailwind), and assets.
4. Manage & Edit: Access your project from the dashboard to preview, edit, or export.

Tech Stack

• Framework: Next.js 16 with App Router
• Database & Backend: Convex
• Authentication: Stack Auth
• AI Providers:
  • Google Gemini (@ai-sdk/google)
  • Groq (@ai-sdk/groq)
  • OpenRouter (@openrouter/ai-sdk-provider)
  • Cerebras (@ai-sdk/cerebras)
• Styling: Tailwind CSS v4
• Components: shadcn/ui
• Runtime & Package Manager: Bun

Getting Started

Prerequisites

• Bun installed on your machine

Installation

1. Clone the repository

2. Install dependencies:

   bun install

3. Set up your environment variables in .env.local (you can start from .env.example or .env.local.example):

   # AI Providers
   GOOGLE_GENERATIVE_AI_API_KEY=your_google_api_key
   GOOGLE_MODEL=gemini-3-flash-preview
   GOOGLE_FALLBACK_MODEL=gemini-2.5-flash
   GROQ_API_KEY=your_groq_api_key
   GROQ_MODEL=moonshotai/kimi-k2-instruct-0905
   GROQ_FALLBACK_MODEL=qwen/qwen3-32b
   OPENROUTER_API_KEY=your_openrouter_api_key
   OPENROUTER_MODEL=openai/gpt-oss-120b
   OPENROUTER_FALLBACK_MODEL=optional_openrouter_fallback_model
   CEREBRAS_API_KEY=your_cerebras_api_key
   CEREBRAS_MODEL=llama-3.3-70b
   CEREBRAS_FALLBACK_MODEL=optional_cerebras_fallback_model
   
   # Convex
   CONVEX_DEPLOYMENT_KEY=your_convex_key # or run bun convex dev
   NEXT_PUBLIC_CONVEX_URL=your_convex_url
   
   # Stack Auth
   NEXT_PUBLIC_STACK_PROJECT_ID=your_stack_project_id
   NEXT_PUBLIC_STACK_PUBLISHABLE_CLIENT_KEY=your_stack_key
   STACK_SECRET_SERVER_KEY=your_stack_secret
   
   # GitHub OAuth (Deployments)
   GITHUB_CLIENT_ID=your_github_oauth_client_id
   GITHUB_CLIENT_SECRET=your_github_oauth_client_secret
   
   # Optional Deployment Integrations
   NETLIFY_CLIENT_ID=your_netlify_oauth_client_id
   NETLIFY_CLIENT_SECRET=your_netlify_oauth_client_secret
   VERCEL_CLIENT_ID=your_vercel_oauth_client_id
   VERCEL_CLIENT_SECRET=your_vercel_oauth_client_secret
   
   # Integration Token Encryption
   INTEGRATION_TOKEN_SECRET=your_32+_char_secret

One-Click Deploy

The editor includes a deploy flow that can create or reuse a GitHub repo and publish the project.

1. Configure deployment integrations in your environment.
2. Open a project in the editor and click Deploy.
3. Connect required accounts when prompted.
4. Choose repo ownership and visibility.
5. Redeploy from the dashboard whenever needed.

Project Settings

Each project has a settings page at /edit/[projectName]/settings that centralizes deployment info and metadata/SEO.

AI Settings & Access Control

• BYOK keys are available to any signed-in user.
• AI admin controls (provider enable/disable, provider defaults, custom models) are restricted to:
  • aditya.mer@somaiya.edu
• AI settings are persisted per user in Convex and synchronized through /api/ai/settings.
• Admin config changes are audit logged in Convex.

4. Initialize Convex:

   bun convex dev

Development

Run the development server:

bun run dev

Open http://localhost:3000 to see the application.

Testing

Run unit tests (Vitest):

bun test

Run type checks:

bun run typecheck

Run lint:

bun run lint

Build the project:

bun run build