Cook from what you already have — AI-powered pantry tracking, recipe matching, and personalized recipe generation.

Live Demo | Architecture | Getting Started

Core

• 🗄️ Pantry Management — add, edit, delete items with quantities, units, and expiry dates
• 📸 AI Fridge Scanner — photograph your fridge/pantry and let Google Gemini identify ingredients automatically
• 📖 Recipe Browser — browse curated recipes matched against your pantry items
• ❤️ Favorites — save recipes you love
• 🛒 Shopping List — track missing ingredients you need to buy

AI-Powered

• 🪄 Magic Recipe Generation — AI generates personalized recipes from your current pantry items
• 🔄 Smart Substitutions — AI suggests ingredient substitutions for missing items
• 📊 Confidence-Based Matching — algorithm scores recipe matches with exact, partial, and fuzzy matching

User Experience

• 👨‍🍳 Cook Mode — distraction-free step-by-step cooking interface with ingredient checklist
• ⚠️ Expiry Alerts — color-coded warnings for expired and soon-to-expire pantry items
• 📈 Dashboard Stats — at-a-glance view of pantry status and makeable recipes
• ⚡ Response Caching — AI responses cached to reduce API calls and improve speed
• 🛡️ Rate Limiting — client-side protection against API abuse

Engineering

• 🔒 Row-Level Security — database-level data isolation per user
• 🧪 154 automated tests with Vitest
• 📝 Full TypeScript (strict mode)
• 🔄 CI/CD pipeline with GitHub Actions
• 🎨 Consistent code style via ESLint + Prettier + Husky

graph TB
    subgraph "Frontend (React SPA)"
        UI[React 19 + TypeScript]
        State[Zustand Stores]
        Cache[Response Cache]
        RL[Rate Limiter]
    end
    
    subgraph "Backend (Supabase)"
        Auth[Supabase Auth]
        DB[(Postgres + RLS)]
        EF[Edge Functions]
    end
    
    subgraph "AI (Google Gemini)"
        Text[Text Generation]
        Multi[Multimodal Vision]
    end
    
    UI --> State
    UI --> Cache
    UI --> RL
    State --> Auth
    State --> DB
    RL --> EF
    EF --> Text
    EF --> Multi
    
    style UI fill:#61dafb
    style DB fill:#336791
    style EF fill:#3ecf8e
    style Text fill:#4285f4
    style Multi fill:#4285f4

• Node.js (LTS)
• A Supabase project
• A Gemini API key configured as a secret in Supabase Edge Functions

git clone https://github.com/Gurehmat/PANTRYCHEFFV3.git
cd PANTRYCHEFFV3
npm install

Create a .env file in the project root:

VITE_SUPABASE_URL=your_supabase_project_url
VITE_SUPABASE_ANON_KEY=your_supabase_anon_key

In the Supabase Dashboard, set the GEMINI_API_KEY secret for Edge Functions, then deploy the functions.

• Apply the SQL in supabase/migrations/ using the Supabase SQL editor or CLI
• In Authentication → URL Configuration, set Site URL and Redirect URLs to match your app, such as your GitHub Pages URL and http://localhost:5173 for local development

Seeding recipes: The app reads recipes from the Supabase recipes table. To populate it, the app uses the file public/data/recipes_with_images.json as the seed source. After you sign in, use the "Seed Database" button (e.g. on the Recipes page); it fetches that JSON, then inserts (and replaces) recipes for your user. Ensure public/data/recipes_with_images.json exists (e.g. copy from src/data/recipes_with_images.json if you keep that as the canonical source).

npm run dev

Open the URL shown in the terminal, usually http://localhost:5173.

npm run build
npm run deploy

The app is configured for GitHub Pages, and deploy publishes the dist folder to the gh-pages branch.

Frontend

• React 19
• TypeScript
• Vite
• React Router
• Zustand
• Tailwind CSS

Backend

• Supabase Auth
• Supabase Postgres
• Row-Level Security
• Supabase Edge Functions

AI

• Google Gemini for text generation
• Google Gemini multimodal vision for ingredient detection

Tooling

• Vitest
• ESLint
• Prettier
• Husky
• GitHub Actions

Run the test suite with:

npm test

Run linting with:

npm run lint

PANTRYCHEFFV3/
├── src/                 # React frontend
├── supabase/
│   ├── functions/       # Edge Functions
│   └── migrations/      # Database schema and policies
├── public/              # Static assets
├── scripts/             # Utility scripts
└── tests/               # Automated tests

• Row-Level Security ensures users can only access their own pantry data
• API secrets are stored securely in Supabase Edge Function secrets
• Rate limiting and caching help reduce abuse and unnecessary API usage

• Barcode scanner for faster pantry entry
• Meal planning calendar
• Nutrition tracking
• Smarter household pantry sharing
• Voice-guided cook mode

This project is for educational and portfolio purposes unless otherwise specified.