🍎 NutriSnap

AI-powered food analysis app. Snap a photo of your ingredients and get personalized recipe suggestions with nutritional breakdowns — powered by Google Gemini 2.0 Flash.

---

Features

• AI Ingredient Detection — Upload a food photo and Gemini identifies all visible ingredients
• Personalized Recipes — Recipes tailored to your diet type, allergies, health goals, and calorie targets
• Nutrition Analysis — Calories, protein, carbs, and fat per recipe
• Health Scoring — Each recipe rated 1–10 relative to your specific fitness goal
• User Accounts — Sign up/sign in via Firebase Authentication
• Save Recipes — Save your favourite recipes to Firestore, accessible across sessions
• Food History — Every analysis is automatically saved for logged-in users
• Mobile Friendly — Responsive design with camera capture support

---

Tech Stack

Layer	Technology
Frontend	React 19 + Vite
Backend	FastAPI (Python)
AI Model	Vertex AI — Gemini 2.0 Flash
Authentication	Firebase Auth (via GCP Identity Platform)
Database	Cloud Firestore
Auth SDK (frontend)	Firebase JS SDK v12
Auth SDK (backend)	firebase-admin

---

Project Structure

nutrisnap/
├── backend/
│   ├── main.py              # FastAPI app — all routes and AI logic
│   ├── requirements.txt     # Python dependencies
│   └── .env                 # GCP_PROJECT_ID (not committed)
├── frontend/
│   ├── src/
│   │   ├── App.jsx          # Main component — upload, auth, results
│   │   ├── App.css          # All styles
│   │   ├── AuthModal.jsx    # Sign in / sign up modal
│   │   ├── firebase.js      # Firebase app initialisation
│   │   └── main.jsx         # React entry point
│   ├── package.json
│   └── vite.config.js
└── README.md

---

Prerequisites

• Node.js v20+
• Python 3.11+
• Google Cloud CLI (gcloud)
• A GCP project with the following enabled:
  • Vertex AI API
  • Cloud Firestore (Native mode)
  • Identity Platform (Email/Password provider)
• A Firebase project linked to your GCP project with a Web App registered

---

Setup

1. Clone the repo

git clone https://github.com/Amogh887/nutrisnap.git
cd nutrisnap

2. Authenticate with Google Cloud

gcloud auth application-default login
gcloud config set project YOUR_PROJECT_ID

3. Backend setup

cd backend
pip install -r requirements.txt

Create backend/.env:

GCP_PROJECT_ID=your-gcp-project-id

4. Frontend setup

cd frontend
npm install

Update frontend/src/firebase.js with your Firebase Web App config (found in Firebase Console → Project Settings → Your Apps).

---

Running the App

Terminal 1 — Backend

cd backend
python main.py

Backend runs at http://localhost:8000

Terminal 2 — Frontend

cd frontend
npm run dev

Frontend runs at http://localhost:5173

---

Mobile Access

To test on your phone (same WiFi network):

1. Find your PC's local IP address:
   • Windows: ipconfig → look for IPv4 Address
   • Mac/Linux: ifconfig or ip addr
2. Open http://YOUR_PC_IP:5173 on your phone

The frontend automatically uses window.location.hostname to route API requests, so no extra config is needed.

---

API Endpoints

Method	Endpoint	Auth	Description
GET	/api/test	None	Health check
POST	/api/analyze-food	Optional	Analyze food image, returns recipes
GET	/api/profile	Required	Get user profile
PUT	/api/profile	Required	Update user profile
GET	/api/preferences	Required	Get dietary preferences
PUT	/api/preferences	Required	Update dietary preferences
GET	/api/saved-recipes	Required	List saved recipes
POST	/api/saved-recipes	Required	Save a recipe
DELETE	/api/saved-recipes/{id}	Required	Delete a saved recipe
GET	/api/food-history	Required	Last 50 food analyses

---

Firestore Schema

users/
  {userId}/
    preferences/    → diet_type, allergies, health_goal, cooking_time,
                      cuisine_preferences, calorie_target, fitness_goal
    saved_recipes/
      {recipeId}/   → full recipe object + saved_at timestamp
    food_history/
      {entryId}/    → detected_ingredients, recipes_generated,
                      analyzed_at, preferences_used

---

Firestore Security Rules

In Firebase Console → Firestore → Rules:

rules_version = '2';
service cloud.firestore {
  match /databases/{database}/documents {
    match /users/{userId}/{document=**} {
      allow read, write: if request.auth != null && request.auth.uid == userId;
    }
  }
}

---

Environment Variables

Variable	Location	Description
GCP_PROJECT_ID	backend/.env	Your Google Cloud project ID

Firebase config values live directly in frontend/src/firebase.js (not secret — Firebase API keys are designed to be public, protected by security rules).

---

Git Workflow

git add .
git commit -m "your message"
git push origin main

The .gitignore excludes .env, venv/, __pycache__/, node_modules/, and dist/.

---

License

MIT