FrostAI

An AI-powered content creation SaaS platform providing tools for article writing, blog title generation, image generation, background removal, object removal, and resume review.

Features

Feature	Description	Plan
Article Writer	Generate comprehensive articles with configurable length	Free (10 uses)
Blog Title Generator	Create SEO-friendly blog titles with category selection	Free (10 uses)
Image Generator	Generate images from text descriptions with style options	Premium
Background Removal	Remove backgrounds from uploaded images	Premium
Object Removal	Remove specified objects from uploaded images	Premium
Resume Review	Upload a PDF resume for AI-powered analysis	Premium
Community	Browse and like publicly shared creations	Premium

Tech Stack

Frontend: React, Vite, Tailwind CSS, React Router, Clerk (auth), Axios, React Hot Toast

Backend: Node.js, Express, Google Gemini (AI text), ClipDrop API (image generation), Cloudinary (image processing), Neon PostgreSQL, Clerk (auth), Multer (file uploads)

Getting Started

Prerequisites

• Node.js 18+
• npm
• Accounts: Clerk, Google AI Studio (Gemini), ClipDrop, Cloudinary, Neon

Installation

# Clone the repository
git clone https://github.com/yourusername/FROST-AI.git
cd FROST-AI
# Install server dependenciescd server
npm install
# Install client dependenciescd ../client
npm install

Environment Variables

Server (server/.env):

DATABASE_URL=your_neon_database_urlCLERK_SECRET_KEY=your_clerk_secret_keyCLERK_PUBLISHABLE_KEY=your_clerk_publishable_keyGEMINI_API_KEY=your_gemini_api_keyCLIPDROP_API_KEY=your_clipdrop_api_keyCLOUDINARY_CLOUD_NAME=your_cloudinary_cloud_nameCLOUDINARY_API_KEY=your_cloudinary_api_keyCLOUDINARY_API_SECRET=your_cloudinary_api_secret

Client (client/.env):

VITE_CLERK_PUBLISHABLE_KEY=your_clerk_publishable_keyVITE_BASE_URL=http://localhost:3000

Running Locally

# Start server (from server/ directory)
npm run server
# Start client (from client/ directory)
npm run dev

The client runs on http://localhost:5173 and proxies API requests to the server on port 3000.

Project Structure

FROST-AI/
├── client/ # React frontend
│ ├── src/
│ │ ├── components/ # Reusable UI components
│ │ ├── pages/ # Route pages
│ │ ├── assets/ # Images, icons, data
│ │ ├── App.jsx # Route definitions
│ │ └── main.jsx # Entry point with ClerkProvider
│ └── vite.config.js
│
└── server/ # Express backend
├── configs/ # DB, Cloudinary, Multer config
├── controllers/ # AI & user route handlers
├── middlewares/ # Auth middleware
├── routes/ # API route definitions
└── server.js # Express entry point

API Endpoints

Method	Endpoint	Description
GET	/	Health check
POST	/api/ai/generate-article	Generate article (free: 10 uses)
POST	/api/ai/generate-blog-title	Generate blog titles (free: 10 uses)
POST	/api/ai/generate-image	Generate image (premium)
POST	/api/ai/remove-image-background	Remove image background (premium)
POST	/api/ai/remove-image-object	Remove object from image (premium)
POST	/api/ai/resume-review	Review PDF resume (premium)
GET	/api/user/get-user-creations	Get user's creations
GET	/api/user/get-published-creations	Get public creations
POST	/api/user/toggle-like-creation	Like/unlike a creation

Deployment

Both client and server are configured for Vercel deployment via vercel.json files in each directory.

License

ISC