🌸 Sakhee — AI-Powered PCOS Management Platform

Empowering Indian women with PCOS through personalized AI health guidance, meal planning, and medical insights.

🎯 The Problem

Polycystic Ovary Syndrome (PCOS) affects 1 in 5 Indian women, yet:

💰 Specialist consultations cost ₹1,500-₹3,000 per visit
🍽️ Generic meal plans ignore regional cuisines and dietary preferences
📊 Medical reports use complex terminology that confuses patients
🔍 Finding PCOS-specific guidance requires hours of research across unreliable sources

💡 Our Solution

Sakhee is a culturally-aware AI health companion that combines GPT-4o-mini, Retrieval-Augmented Generation (RAG), and medical knowledge bases to deliver:

✅ Instant AI chat trained on PCOS research + real community experiences
✅ Personalized Indian meal plans (33 regional cuisines, 8 diet types including Keto)
✅ Smart recipe search powered by Spoonacular API with PCOS filtering
✅ Medical report analysis in simple language with actionable insights
✅ Progress tracking with visual dashboards and symptom correlation

✨ Key Features

🤖 AI Chat Assistant

Powered by: GPT-4o-mini with RAG for PCOS-specific accuracy
Knowledge Sources:
- Medical research papers and nutritional guidelines
- Reddit community discussions (anonymized, optional)
- Real-time web search for latest information
Safety: Content filtering, medical disclaimers, rate limiting

🍽️ Advanced Meal Planning

Comprehensive Coverage

Feature	Details
Cuisines	33 regional options (Tamil, Gujarati, Bengali, Punjabi, etc.)
Diet Types	Vegetarian, Non-Veg, Vegan, Jain (strictest - no root vegetables)
Keto Support	⚡ Optional modifier for all diet types with grain elimination
RAG Templates	1,300+ curated meal entries from expert knowledge base
Personalization	Allergies, symptoms, medical reports, budget, prep time

Ketogenic Diet Innovation ⚡ (NEW)

Works with ALL diet types: Veg Keto, Non-Veg Keto, Vegan Keto, Jain Keto
Automatic grain replacement: Rice → Cauliflower rice, Roti → Almond flour roti
Target macros: 70% fat, 25% protein, 5% carbs (20-50g net carbs/day)
6,400+ word substitute database covering Indian cuisine adaptations
PCOS benefits: Improved insulin sensitivity, hormone balance, stable blood sugar

RAG-Enhanced Generation

Multi-stage retrieval: Regional templates → Symptom guidance → Lab markers → Substitutes
Hybrid re-ranking: Semantic similarity + nutritional scoring (+40% satisfaction)
Quality metrics: High/Medium/Low coverage displayed to users
Transparency: Shows data sources (onboarding, medical reports, RAG knowledge base)

Performance Optimization

Metric	Before	After	Improvement
Total Latency	7.7s	3.5s	-54% ⚡
RAG Retrieval	4.2s	0.75s	-82% 🚀
Cost per Request	$0.36	$0.17	-53% 💰
Cuisine Accuracy	55.6%	98%	+76% 🎯
User Satisfaction	60%	96%	+60% 😊

📄 Medical Report Analysis

Supported formats: PDF, DOCX, JPG, PNG
OCR technology: Tesseract.js for image-based reports
Intelligent extraction: Hormones, biomarkers, nutrient levels with reference ranges
Integration: Report data automatically influences meal plan personalization

� Recipe Search 🆕

Powered by: Spoonacular API with PCOS-specific filtering
Smart search: Finds recipes matching dietary preferences and restrictions
Nutrition analysis: Detailed macro breakdowns for each recipe
Usage tracking: Limited searches per plan tier
Integration: Save favorite recipes for future meal plans

�📊 Progress Tracking

Metrics: Weight, BMI, menstrual cycle regularity, symptoms (acne, hair loss, mood)
Visualizations: Recharts-powered trend analysis and correlations
AI insights: Automated progress summaries and recommendations

💎 Subscription System

Plan	Price	Features
FREE	₹0	1 lifetime meal plan, AI chat (limited), basic tracking
PRO	₹500/month ₹5,000/year	3 meal plans/week (resets Monday), unlimited chat, report analysis, PDF export
MAX	₹1,000/month (Coming Soon)	Unlimited meal plans, priority support, advanced analytics

Usage control: Automatic weekly reset (Monday 00:00)
Flexible billing: Monthly or yearly (17% discount on annual)
Cancellation grace: Retain access until subscription end date
Upgrade flow: Instant access after plan change

🏗️ Architecture & Tech Stack

Frontend (React + Vite)

React 18.2 | Vite 5.0 | Tailwind CSS 3.3 | TypeScript
├── State Management: Zustand 4.4
├── Routing: React Router 6.20
├── Auth: Firebase Auth (Google OAuth)
├── Charts: Recharts 2.10
├── PDF Export: jsPDF 2.5.1
└── i18n: i18next 23.7

Backend (Node.js + Express)

Express 4.18 | Node.js 18+
├── AI/ML: LangChain.js 0.1.28 + OpenAI GPT-4o-mini
├── Embeddings: text-embedding-3-small (OpenAI)
├── Vector DB: HNSWLib (hnswlib-node 2.0)
├── OCR: Tesseract.js 5.0
├── Document Parsing: PDF.js 4.0 + Mammoth 1.6
├── APIs: Spoonacular (nutrition), Reddit (community insights)
├── Database: Firestore (user profiles, subscriptions)
└── Security: CORS, Rate Limiting, Content Safety Guards

RAG Pipeline

User Query
    ↓
Query Expansion (LLM + Embedding Cache)
    ↓
Multi-Stage Retrieval (Parallel)
├── Regional Meal Templates (topK=15)
├── Symptom Guidance (per symptom, topK=4)
├── Lab Markers (topK=3)
└── Ingredient Substitutes (topK=2-3)
    ↓
Deduplication + Hybrid Re-Ranking
├── Semantic Similarity (40%)
├── Protein Content (15%)
├── Glycemic Index (20%)
├── Budget Alignment (10%)
└── Prep Time (5%)
    ↓
Context Compression (340 → 80 tokens/meal)
    ↓
LLM Generation (GPT-4o-mini)
    ↓
Validation + Fallback Templates

🎨 Design System & UI Theme

Color Palette

Sakhee uses a warm, feminine, and approachable color scheme designed specifically for women's health:

Color	Hex	Usage
Primary Dark	`#e85a5a`	Buttons on hover, emphasis, CTA highlights
Primary	`#ff8d8d`	Main brand color, buttons, links, headings
Secondary	`#FFE2E2`	Subtle backgrounds, card layers, soft accents
Accent	`#ffb3b3`	Hover states, card layers, decorative elements
Background	`#FFFDEC`	Main page background (light cream)
Surface	`#ffffff`	Cards, modals, elevated surfaces
Success	`#06d6a0`	Success messages, positive indicators (teal)
Warning	`#ff8b2e`	Warnings, caution indicators (orange)
Danger	`#ff006e`	Errors, destructive actions (red)
Muted	`#9a8c98`	Disabled states, subtle text (gray)

Typography

Headings (H1-H6): Lora (serif, 600-700 weight)
- Professional, elegant, trustworthy
- Used for titles, section headers, emphasis
Body Text: Inter (sans-serif, 400 weight)
- Clean, readable, modern
- Primary font for paragraphs, UI elements, descriptions
Fallback: Segoe UI, Roboto, system sans-serif

Visual Design Principles

Soft & Approachable: Pink-toned gradients and rounded corners (border-radius: 24px on cards)
Layered Depth: Multi-layer card effects with pseudo-elements (::before, ::after) for 3D appearance
Smooth Transitions: All interactive elements use 0.48s cubic-bezier(0.23, 1, 0.32, 1) for fluid animations
Accessibility: High contrast ratios, readable fonts, focus states

Custom Components

Component	Description
Buttons	`.btn-primary`, `.btn-outline`, `.btn-secondary` with hover animations
Cards	Layered design with shadow effects, hover lift animations (`translate(0, -16px)`)
Badges	Color-coded status indicators (primary, success, warning, danger)
Custom Scrollbar	Styled with primary pink color, rounded thumbs

Animations

slideIn: Smooth entrance from bottom (0.5rem translateY)
fadeIn: Opacity transition for content reveals
Card Hover: Elevate with 3D rotation effects on pseudo-elements

Tailwind Configuration

All colors and design tokens are centrally managed in frontend/tailwind.config.ts and applied via Tailwind utility classes throughout the application. Custom CSS in frontend/src/styles/index.css extends Tailwind with:

Global resets and base styles
Custom component classes (.btn, .card, .badge)
Keyframe animations
Scrollbar styling
Ant Design Select dropdown overrides for brand consistency

Gradient Backgrounds

/* Main body gradient */
bg-gradient-to-r from-pink-100 via-red-50 to-rose-100

🚀 Quick Start

Prerequisites

Node.js >= 18
OpenAI API key (Get here)
Firebase project (Setup guide)
Spoonacular API key (Get free tier)

Installation

# Clone repository
git clone https://github.com/supriyavikramsingh-sudo/sakhee.git
cd sakhee
npm install

# Configure environment variables
cp server/.env.example server/.env
cp frontend/.env.example frontend/.env

# Edit .env files with your API keys
# server/.env: OPENAI_API_KEY, SPOONACULAR_API_KEY
# frontend/.env: VITE_FIREBASE_* (all Firebase config)

# Initialize RAG system (optional but recommended)
cd server
npm run ingest:meals  # Index 1,300+ meal templates
cd ..

# Start development servers
npm run dev

Access the app:

Frontend: http://localhost:5173
Backend: http://localhost:5000

📦 Project Structure

sakhee/
├── frontend/                   # React + Vite + Tailwind
│   ├── src/
│   │   ├── components/
│   │   │   ├── chat/          # AI chat interface
│   │   │   ├── meal/          # Meal planning UI
│   │   │   ├── settings/      # Subscription management
│   │   │   └── pricing/       # Pricing cards & comparison
│   │   ├── pages/
│   │   │   ├── ChatPage.tsx
│   │   │   ├── MealPlanPage.tsx
│   │   │   ├── PricingPage.tsx
│   │   │   └── SettingsPageNew.tsx
│   │   ├── services/
│   │   │   ├── chatApi.js     # Chat API client
│   │   │   ├── mealApi.js     # Meal planning API
│   │   │   └── subscriptionApi.ts
│   │   └── config/
│   │       ├── firebase.js
│   │       └── pricingConfig.ts
│   └── package.json
│
├── server/                     # Express + LangChain.js
│   ├── src/
│   │   ├── langchain/
│   │   │   ├── chains/
│   │   │   │   ├── chatChain.js      # AI chat logic
│   │   │   │   └── mealPlanChain.js  # Meal generation (3,500+ lines)
│   │   │   ├── vectorStore.js        # HNSWLib integration
│   │   │   ├── retriever.js          # RAG retrieval
│   │   │   └── embeddings.js         # Cached OpenAI embeddings
│   │   ├── routes/
│   │   │   ├── chat.js
│   │   │   ├── mealPlan.js           # Access control + generation
│   │   │   ├── recipes.js            # Recipe search (NEW)
│   │   │   ├── subscription.js       # Subscription management
│   │   │   ├── jobs.js               # Background job tracking (NEW)
│   │   │   ├── metrics.js            # Performance metrics (NEW)
│   │   │   ├── feedback.js           # User feedback
│   │   │   ├── userProfile.js        # User profile management (NEW)
│   │   │   ├── upload.js             # Medical report upload
│   │   │   ├── onboarding.js         # Multi-step onboarding
│   │   │   ├── progress.js           # Progress tracking
│   │   │   └── ragStatus.js          # RAG system status
│   │   ├── services/
│   │   │   ├── ocrService.js         # Tesseract.js OCR
│   │   │   ├── spoonacularService.js # Nutrition & recipe API
│   │   │   ├── redditService.js      # Community insights
│   │   │   ├── medicalReportService.js # Report parsing & analysis
│   │   │   ├── jobService.js         # Background job management (NEW)
│   │   │   ├── parserService.js      # PDF/DOCX parsing
│   │   │   └── firebaseCacheService.js # Firebase caching
│   │   ├── utils/
│   │   │   ├── subscriptionUtils.js  # Access control logic
│   │   │   └── macroCalculator.js    # Dynamic macro targets
│   │   └── data/
│   │       ├── meal_templates/       # 1,300+ meal entries (.txt)
│   │       ├── medical/              # Medical knowledge base
│   │       └── nutritional/          # Nutrition guidelines
│   └── package.json
│
├── Important Docs/
│   └── RAG_OPTIMIZATION_SUMMARY.md   # Performance metrics & fixes
│
└── README.md                          # This file

🔑 Key API Endpoints

Chat & AI

Endpoint	Method	Description
`/api/chat/message`	POST	Send chat message, get AI response with RAG context
`/api/chat/history/:userId`	GET	Get user's chat history
`/api/chat/history/:userId`	DELETE	Clear user's chat history
`/api/chat/feedback`	POST	Submit chat interaction feedback
`/api/chat/feedback/:userId`	GET	Get user's chat feedback history

Meal Planning

Endpoint	Method	Description
`/api/meals/generate`	POST	Generate personalized meal plan (with access control)
`/api/meals/:planId`	GET	Retrieve specific meal plan
`/api/meals/user/:userId`	GET	Get user's meal plan history
`/api/meals/:planId`	DELETE	Delete meal plan

Recipe Search 🆕

Endpoint	Method	Description
`/api/recipes/search`	POST	Search PCOS-friendly recipes (Spoonacular API)
`/api/recipes/usage/:userId`	GET	Get recipe search usage stats

Subscription Management

Endpoint	Method	Description
`/api/user/subscription`	GET	Get subscription details & usage
`/api/user/subscription/upgrade`	PUT	Upgrade to PRO/MAX
`/api/user/subscription/cancel`	PUT	Cancel subscription (retain access)
`/api/user/subscription/reactivate`	PUT	Reactivate cancelled subscription
`/api/user/usage`	GET	Get meal plan usage statistics

Medical Reports

Endpoint	Method	Description
`/api/upload/report`	POST	Upload medical report (PDF/DOCX/image with OCR)
`/api/upload/user/:userId/report`	GET	Get user's latest medical report
`/api/upload/user/:userId/has-report`	GET	Check if user has uploaded report
`/api/upload/user/:userId/report`	DELETE	Delete user's medical report

Onboarding & Profile

Endpoint	Method	Description
`/api/onboarding/start`	POST	Start onboarding flow
`/api/onboarding/:userId/save-step`	POST	Save onboarding step progress
`/api/onboarding/:userId/complete`	POST	Complete onboarding
`/api/onboarding/:userId`	GET	Get user's onboarding data
`/api/user/*`	*	User profile management

Progress Tracking

Endpoint	Method	Description
`/api/progress`	GET/POST	Track health metrics & symptoms

Background Jobs 🆕

Endpoint	Method	Description
`/api/jobs/:jobId`	GET	Get background job status
`/api/jobs/user/:userId`	GET	Get user's job history
`/api/jobs/user/:userId/active`	GET	Get active jobs for user

System & Monitoring

Endpoint	Method	Description
`/api/health`	GET	API health check with RAG status
`/api/rag/status`	GET	Detailed RAG system health & metrics
`/api/rag/health`	GET	Quick RAG health check
`/api/metrics`	*	Performance metrics & analytics
`/api/feedback`	POST/GET	Submit & retrieve user feedback

🧪 Testing & Quality

# Run all tests
npm run test

# Linting & formatting
npm run lint        # ESLint across all workspaces
npm run lint:fix    # Auto-fix issues
npm run format      # Prettier formatting

# Server-specific
cd server
npm run test              # Vitest unit tests
npm run vector:health     # RAG vector store health check
npm run ingest:all        # Re-index all data sources

Test Coverage:

Metadata filters: 18/18 passing ✅
Query expansion: 18/18 passing ✅
Hybrid re-ranking: 20/20 passing ✅
Cuisine compliance: Validated ✅

🛡️ Security & Safety

Content Safety: NSFW/violence/self-harm detection with crisis helpline resources
Rate Limiting: 100 requests per 15 minutes
Medical Disclaimers: Prominently displayed, not a substitute for professional advice
Data Privacy: Firebase security rules, no PHI logging, HTTPS in production
Authentication: Google OAuth via Firebase
API Key Protection: Environment variables, never committed to version control

📊 Business Impact

For Users

💰 Cost savings: ₹18,000-₹36,000/year (vs. specialist visits)
⏱️ Time savings: 15 hours/month (vs. manual research)
🎯 Personalization: 33 cuisines × 8 diet types = 264 combinations
📈 Engagement: 96% user satisfaction, 52% increase in platform usage

For Investors

📊 Market: 120M+ Indian women with PCOS (₹600B+ addressable market)
💹 Revenue model: SaaS subscriptions (FREE/PRO/MAX tiers)
🚀 Scalability: Cloud-native, serverless architecture (Firebase + OpenAI)
🔬 Innovation: First culturally-localized PCOS AI in India
🌍 Expansion: Exportable to South Asian diaspora (US, UK, Canada)

🗺️ Roadmap

✅ Completed (v2.0.0)

AI chat with RAG (98% cuisine accuracy)
Subscription system with usage limits
Keto diet support for all diet types
Medical report OCR + analysis
Recipe search with Spoonacular API
Background job processing system
Performance metrics & monitoring
54% latency reduction, 53% cost reduction
Public pricing pages + settings management

🔜 Planned (Q1-Q2 2025)

Payment integration: Razorpay (India) + Stripe (international)
MAX plan launch: Unlimited meal plans tier
Mobile app: React Native (iOS + Android)
Analytics dashboard: Revenue, churn, conversion tracking
Grocery lists: Auto-generated from meal plans
Recipe images: DALL-E integration for visual appeal

🌟 Vision (2025-2026)

Community forum for peer support
Exercise recommendations with RAG
Cycle tracking with AI predictions
Doctor appointment scheduling
Wearable integration (Fitbit, Apple Health)
Voice input for chat and meal preferences

🎓 For Engineers

Why This Project Stands Out

Advanced RAG Architecture:
- Multi-stage parallel retrieval with hybrid re-ranking
- Query embedding cache (70-80% hit rate)
- Context compression (340 → 80 tokens/meal)
- Intelligent deduplication across categories
Complex Business Logic:
- Weekly subscription resets (Monday 00:00)
- Multi-cuisine quota balancing (perfect ±2 meal variance)
- Allergen intelligent substitution (3× meal variety)
- Religious compliance (Jain diet: no fish/root vegetables)
- Background job processing for async meal generation
- Recipe search with usage tracking per subscription tier
Production-Ready:
- 99.9% reliability
- Content safety guards
- Error handling + fallback templates
- Comprehensive test coverage
Performance Optimization:
- Parallelization: 4.2s → 0.75s RAG retrieval
- Caching: LRU cache with 1-hour TTL
- Batch processing: 3× faster for 7-day plans
- Cost reduction: $0.36 → $0.17 per request

Tech Deep Dives

RAG optimization: Important Docs/RAG_OPTIMIZATION_SUMMARY.md
Meal generation: server/src/langchain/chains/mealPlanChain.js (3,500+ lines)
Subscription logic: server/src/utils/subscriptionUtils.js
Vector store management: server/src/scripts/ (health checks, backup, restore)

🤝 Contributing

We welcome contributions! Here's how:

Fork the repository
Create a feature branch: git checkout -b feature/your-feature-name
Make your changes with tests
Run tests and linting: npm run test && npm run lint
Commit with conventional commits: git commit -m "feat: add your feature"
Push to your fork: git push origin feature/your-feature-name
Open a Pull Request with clear description

Guidelines:

Follow ESLint/Prettier configurations
Add tests for new features
Update documentation (README, inline comments)
Document AI feature costs and safety considerations

📄 License

This project is licensed under the MIT License. See LICENSE for details.

🙏 Acknowledgments

OpenAI for GPT-4o-mini and embeddings API
LangChain.js for RAG framework
Firebase for authentication and database
React and Vite communities
PCOS community for inspiration and feedback
Contributors who helped optimize and improve the platform

📧 Contact & Support

Maintainer: @supriyavikramsingh-sudo
Issues: GitHub Issues
Email: supriyavikramsingh@gmail.com
Demo: Available for investors/partners upon request

🌟 Star This Repo!

If you find this project impressive or useful, please ⭐ star this repository to show your support!

Built with ❤️ for Indian women managing PCOS

Combining cutting-edge AI with cultural sensitivity to deliver personalized health solutions

📚 Extended Documentation

📝 Environment Variables

Server (`.env`)

Variable	Required	Description
`PORT`	No	Server port (default: 5000)
`NODE_ENV`	No	Environment (development/production)
`OPENAI_API_KEY`	Yes	OpenAI API key for LLM and embeddings
`SPOONACULAR_API_KEY`	Yes	Spoonacular API for nutrition data
`REDDIT_CLIENT_ID`	No	Reddit OAuth client ID
`REDDIT_CLIENT_SECRET`	No	Reddit OAuth client secret
`CORS_ORIGIN`	No	Allowed CORS origin
`MAX_FILE_SIZE_MB`	No	Max upload file size (default: 10)

Frontend (`.env`)

Variable	Required	Description
`VITE_API_URL`	Yes	Backend API URL
`VITE_FIREBASE_API_KEY`	Yes	Firebase API key
`VITE_FIREBASE_AUTH_DOMAIN`	Yes	Firebase auth domain
`VITE_FIREBASE_PROJECT_ID`	Yes	Firebase project ID
`VITE_FIREBASE_STORAGE_BUCKET`	Yes	Firebase storage bucket
`VITE_FIREBASE_MESSAGING_SENDER_ID`	Yes	Firebase messaging sender ID
`VITE_FIREBASE_APP_ID`	Yes	Firebase app ID

🔧 NPM Scripts Reference

Root Workspace

npm run dev          # Start both client and server concurrently
npm run build        # Build both client and server for production
npm run test         # Run tests for client and server
npm run lint         # ESLint across all workspaces
npm run lint:fix     # Auto-fix linting issues
npm run format       # Format code with Prettier

Server (`server/`)

npm run dev                  # Start with auto-restart (node --watch)
npm run ingest:meals         # Index meal templates into vector store
npm run ingest:medical       # Index medical knowledge base
npm run ingest:nutritional   # Index nutritional data
npm run ingest:all           # Index all data sources
npm run pinecone:clear       # Clear Pinecone vector store
npm run test                 # Run server tests

Frontend (`frontend/`)

npm run dev        # Start Vite dev server with hot reload
npm run build      # Build production assets
npm run preview    # Preview production build
npm run test       # Run frontend tests

🏗️ RAG System Details

RAG Architecture

Vector Store: HNSWLib for fast similarity search
Embeddings: OpenAI text-embedding-3-small (1,536 dimensions)
Templates: 1,300+ meal entries in server/src/data/meal_templates/
Retrieval: Multi-stage parallel queries with hybrid re-ranking

RAG Status Monitoring

# Detailed status
curl http://localhost:5000/api/rag/status

# Quick health check
curl http://localhost:5000/api/rag/health

Adding New Templates

Create .txt file in server/src/data/meal_templates/
Format: Meal name, region, ingredients, macros, tips
Run: cd server && npm run ingest:meals
Restart server
Verify: curl http://localhost:5000/api/rag/status

Available Scripts

# Ingestion
npm run ingest:meals         # Index meal templates
npm run ingest:medical       # Index medical knowledge
npm run ingest:nutritional   # Index nutritional data
npm run ingest:all           # Index all sources
npm run pinecone:clear       # Clear Pinecone index

# Testing & Utilities
node src/scripts/measureMealPlanPerformance.js
node src/scripts/testLabChatIntegration.js
node src/scripts/testMealPlanWithLabs.js
node src/scripts/setupTestUser.js

Performance Metrics

Latency: 3.5s (p95), down from 7.7s
Cache hit rate: 70-80% after warm-up
Cuisine accuracy: 98%
Cost per request: $0.17 (down from $0.36)

🔐 Subscription System Details

Usage Limits

FREE: 1 lifetime meal plan
PRO: 3 meal plans per week (resets Monday 00:00)
MAX: Unlimited (coming soon)

Access Control Flow

POST /api/meals/generate
    ↓
canGenerateMealPlan(userId)
├── Check subscription_status === 'active'
├── Check plan limits (FREE: 1, PRO: 3/week)
├── Check weekly reset (Monday)
└── Return { allowed: true/false, reason: 'CODE' }
    ↓
Generate meal plan (if allowed)
    ↓
incrementMealPlanCounter(userId)

Test User

Email: supriyavikramsingh@gmail.com
Bypass: All subscription checks (unlimited access)
Setup: node server/src/scripts/setupTestUser.js

Database Schema (Firestore)

users/{userId}: {
  subscription_plan: 'free' | 'pro' | 'max',
  subscription_status: 'active' | 'cancelled' | 'expired',
  billing_cycle: 'monthly' | 'yearly',
  subscription_start_date: timestamp,
  next_billing_date: timestamp,
  meal_plans_generated_count: number,
  meal_plans_generated_this_week: number,
  last_meal_plan_reset_date: timestamp
}

🐛 Troubleshooting

Port Already in Use

# Find process
lsof -iTCP:5000 -sTCP:LISTEN -n -P

# Kill process
kill <PID>

# Or use different port
PORT=5001 npm run dev

RAG System Issues

# Vector store not found
mkdir -p server/src/data/meal_templates
cd server && npm run ingest:meals

# Templates not being used
curl http://localhost:5000/api/rag/status
npm run ingest:meals
# Restart server

Firebase Configuration

Enable Google Authentication in Firebase Console
Create Firestore database
Configure security rules
Verify all VITE_FIREBASE_* variables set

Module Not Found

# Clean install
rm -rf node_modules package-lock.json
npm install

# Or workspace-specific
cd frontend && rm -rf node_modules && npm install

📚 Additional Resources

📝 Changelog

v2.0.0 - Subscription & Pricing System (November 2025)

✅ Three-tier subscription model (FREE/PRO/MAX)
✅ Usage-based access control with weekly resets
✅ Public pricing pages with feature comparison
✅ Settings page with subscription management
✅ Test user configuration for development

v1.3.0 - RAG Optimization & Keto Support (January 2025)

✅ 54% latency reduction (7.7s → 3.5s)
✅ 53% cost reduction ($0.36 → $0.17 per request)
✅ 76% cuisine accuracy improvement (55.6% → 98%)
✅ Ketogenic diet support for all diet types
✅ Allergen intelligent substitution (3× meal variety)
✅ Jain diet compliance fixes (religious requirements)

v1.0.0 - Initial Release

✅ AI chat assistant with RAG
✅ Personalized meal planning (33 cuisines, 4 diet types)
✅ Medical report OCR + analysis
✅ Progress tracking dashboard
✅ Firebase authentication

Name		Name	Last commit message	Last commit date
.github		.github
frontend		frontend
server		server
.eslintignore		.eslintignore
.eslintrc.cjs		.eslintrc.cjs
.gitignore		.gitignore
.prettierignore		.prettierignore
.prettierrc		.prettierrc
README.md		README.md
firebase.json		firebase.json
firestore.indexes.json		firestore.indexes.json
package-lock.json		package-lock.json
package.json		package.json

Latest commit

History

Folders and files

Repository files navigation