Cook Helper - Smart Cooking Assistant Application

A smart cooking assistant application built with React, TypeScript, and Supabase that helps users manage ingredients, discover recipes, and get AI-powered recommendations.

Features

🥘 Core Features

• Ingredient Management: Track your available ingredients with quantities, units, and expiry dates
• Recipe Collection: Create, edit, and manage your personal recipe collection
• Weekly Meal Planning: Plan your meals for the week and generate shopping lists

🤖 AI-Powered Features

• AI Recipe Suggestions: Get personalized recipe suggestions using Google's Gemini AI
• Dataset Recommendations: Discover popular recipes from a curated dataset
• RAG (Retrieval-Augmented Generation): Advanced semantic search using vector embeddings

🔍 Smart Search

• Semantic Search: Find recipes based on meaning, not just keywords
• Vector Similarity: Uses pgvector for efficient similarity search
• Background Sync: Embeddings are generated and updated automatically via cron jobs

Technology Stack

• Frontend: React 18, TypeScript, Tailwind CSS
• Backend: Supabase (PostgreSQL + Edge Functions)
• AI/ML: Google Gemini AI, pgvector for embeddings
• Icons: Lucide React
• Build Tool: Vite

Database Schema

Core Tables

• ingredients - User's ingredient inventory
• recipes - User's personal recipes
• recipe_ingredients - Recipe ingredient relationships
• weekly_plans - Weekly meal planning
• daily_meals - Daily meal assignments

Dataset & AI Tables

• dataset_recipes - Curated recipe dataset
• recipe_embeddings - Vector embeddings for semantic search

Setup Instructions

Prerequisites

• Node.js 18+
• Supabase account
• Google AI API key (for Gemini)

Environment Variables

Create a .env file with:

VITE_SUPABASE_URL=your_supabase_url
VITE_SUPABASE_ANON_KEY=your_supabase_anon_key
VITE_GOOGLE_AI_API_KEY=your_google_ai_api_key

Installation

# Install dependencies
npm install

# Run development server
npm run dev

# Build for production
npm run build

Database Setup

The application includes Supabase migrations that will:

1. Create all necessary tables
2. Set up Row Level Security (RLS) policies
3. Enable pgvector extension
4. Create vector similarity functions

Background Embedding Sync

The application uses a background cron job to generate and sync recipe embeddings:

Edge Function: sync-embeddings

• Processes recipes without embeddings
• Generates vector embeddings for semantic search
• Runs automatically via cron job
• Can be triggered manually for testing

Cron Job Setup

Set up a cron job to call the sync function regularly:

# Run every 6 hours
0 */6 * * * curl -X POST "https://your-project.supabase.co/functions/v1/sync-embeddings" \
  -H "Authorization: Bearer YOUR_ANON_KEY"

Features Overview

1. Ingredient Manager

• Add ingredients with quantities, units, and categories
• Track expiry dates
• Organize by categories (vegetables, meat, dairy, etc.)

2. Recipe Manager

• Create detailed recipes with ingredients and instructions
• Set cooking times, difficulty levels, and serving sizes
• Add tags for easy categorization

3. AI Recipe Suggestions

• Uses Google Gemini AI to generate recipes based on available ingredients
• Provides Indonesian-style recipes with local ingredients
• Saves generated recipes to your collection

4. Dataset Recommendations

• Recommends recipes from a curated dataset of popular recipes
• Filters by ingredient availability and popularity
• Shows match scores and reasons for recommendations

5. RAG Recipe Intelligence

• Advanced semantic search using vector embeddings
• Understands context and meaning beyond keyword matching
• Background processing ensures fast response times
• Confidence scoring for recommendation quality

6. Weekly Meal Planner

• Plan meals for the entire week
• Automatically generate shopping lists
• Track what ingredients you already have

Architecture Highlights

Vector Search Implementation

• Uses pgvector extension for efficient similarity search
• 384-dimensional embeddings for recipe content
• Cosine similarity for finding related recipes
• Background sync prevents user-facing delays

Row Level Security

• Comprehensive RLS policies for data protection
• Users can only access their own data
• Service role for background operations
• Public access for dataset recipes

Performance Optimizations

• Background embedding generation
• Efficient vector indexing
• Caching strategies for embeddings
• Fallback mechanisms for reliability

Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests if applicable
5. Submit a pull request

License

This project is licensed under the MIT License.