An AI-powered mental health platform — mood tracking, clinical assessments, an AI companion, and verified doctor booking, in one calm, beautiful product.
Live Demo · Features · Architecture · Tech Stack · Wiki
| Wellness Dashboard | Mood Tracker |
|---|---|
 |
 |
| Clinical Assessment | Resource Library |
|---|---|
 |
 |
BetterMind is a full-stack mental health platform built with the Next.js App Router. It brings together the day-to-day tools people actually need for mental wellness — daily mood logging, clinically-validated assessments (PHQ-9 & GAD-7), an AI chat companion, and appointment booking with verified mental-health professionals — behind a single, calming, accessibility-minded interface.
It supports three distinct roles (patient, doctor, admin), each with its own dashboard and permissions, all secured by a custom JWT auth layer with optional TOTP two-factor authentication, AES-encrypted sensitive data, and rate-limited APIs.
Any visitor who isn't signed in is automatically in Guest Mode — no button to click, no opt-in required. The full platform is immediately browsable with realistic demo data: mood trends, assessment history, AI conversations, doctor listings, and appointments.
The moment a guest tries to write data (save a mood, book an appointment, message the AI), a modal invites them to create an account — so the experience is fully explorable but nothing leaks into the database.
| Feature | Description |
|---|---|
| Mood Tracking | Log daily mood (1–10) with activities & notes; visualized as interactive trend charts |
| Clinical Assessments | Standardized PHQ-9 (depression) and GAD-7 (anxiety) with automatic severity scoring and a question-by-question history |
| AI Companion | Chat powered by OpenAI gpt-4o-mini, with the user's recent mood & assessment scores injected as context for personalized, safety-aware responses |
| Insightful Dashboard | Mood trends, assessment trends, activity-impact analysis, and mood↔severity correlation, all in light, animated Recharts visualizations |
| Doctor Discovery & Booking | Browse verified professionals by specialty, view ratings & reviews, and book real time-slots generated from each doctor's working hours |
| Google Calendar Sync | Appointments optionally sync to the doctor's Google Calendar via OAuth |
| Resource Library | 100+ curated articles, videos & tools, filterable by tag/category, with personalized recommendations based on your assessment and mood data |
- Verification workflow (pending → approved/rejected with reason)
- Profile editor with education, experience, and a Mon–Sun working-hours grid
- Dashboard with patient stats, ratings, and upcoming appointments
- Patient reviews & ratings (one per completed appointment)
- Doctor verification console (approve / reject with reason)
- Tabbed views: pending, verified, rejected
┌──────────────────────────────────────────────────────────────┐
│ Next.js 14 App Router │
│ │
│ Client Components (RSC-aware) API Route Handlers │
│ ┌────────────────────────┐ ┌────────────────────────┐ │
│ │ Pages: dashboard, mood, │ fetch │ /api/auth/* (JWT, 2FA) │ │
│ │ assessment, chat, │ ───────► │ /api/mood, /assessment │ │
│ │ doctors, appointments… │ │ /api/chat (OpenAI) │ │
│ │ │ ◄─────── │ /api/doctors, /appts │ │
│ │ AuthContext (+ Guest) │ JSON │ /api/doctor/*, /admin/* │ │
│ └────────────────────────┘ └───────────┬─────────────┘ │
└───────────────────────────────────────────────────┼──────────────┘
│
┌───────────────────────────────────┼───────────────────┐
│ │ │ │
┌─────▼─────┐ ┌──────▼──────┐ ┌──────▼──────┐ ┌────────▼────────┐
│ MongoDB │ │ OpenAI │ │ Google APIs │ │ otplib (TOTP) │
│ (driver) │ │ gpt-4o-mini │ │ Calendar │ │ + AES crypto │
└────────────┘ └─────────────┘ └─────────────┘ └─────────────────┘
Key design decisions
- Custom JWT auth (httpOnly cookie) over a heavier auth framework — full control over the 2FA challenge flow and role gating.
- Server-side auth verification with a short-lived in-memory cache to avoid hammering the DB on every request.
- Graceful degradation everywhere: AI chat falls back to keyword matching without an API key; Calendar sync no-ops without OAuth tokens; the app boots even if optional env vars are missing.
- Guest mode as a first-class client concern — demo data lives entirely client-side and write actions are intercepted by a single
requireRealUser()gate.
| Layer | Technologies |
|---|---|
| Framework | Next.js 14 (App Router, Route Handlers, Server/Client Components) |
| UI | React 18, Tailwind CSS 3, custom design-token system |
| Animation | Framer Motion 12 (layout animations, AnimatePresence, shared-layout nav) |
| Data Viz | Recharts 3 (area, line, bar, scatter charts) |
| Database | MongoDB 6 (official driver) |
| Auth | JWT (jsonwebtoken), httpOnly cookies, bcryptjs, otplib TOTP 2FA, qrcode |
| AI | OpenAI SDK (gpt-4o-mini) with context injection + keyword fallback |
| Integrations | Google APIs (googleapis) — Calendar sync via OAuth |
| Security | AES encryption for sensitive fields, in-memory rate limiting, role-based access control |
| Icons | lucide-react |
- JWT authentication stored in httpOnly cookies, verified server-side on every protected route.
- Two-Factor Authentication (TOTP) — full enable/verify/disable flow with QR provisioning and one-time recovery codes; JWT issuance is gated on the 2FA challenge at login.
- AES encryption for sensitive data at rest.
- Rate limiting — login, registration (5/15min), chat (30/min), and assessments (20/hr) are throttled via a reusable limiter factory.
- Role-based access control — patient / doctor / admin separation enforced both client- and server-side.
- Doctor verification — professionals cannot accept appointments until an admin approves their credentials.
- Node.js 18+
- A MongoDB instance (local or Atlas)
- (Optional) OpenAI API key, Google OAuth credentials
# 1. Clone
git clone https://github.com/harshcode1/BetterMind.git
cd BetterMind
# 2. Install
npm install
# 3. Configure environment — create .env.local in the project root
# (see Environment Variables section below)
# 4. Seed demo doctors (optional but recommended for a full demo)
node scripts/seed-doctors.mjs
# 5. Run
npm run devOpen http://localhost:3000 — the platform loads in demo mode automatically if you're not signed in.
Create a .env.local file in the project root:
# Required
MONGODB_URI=mongodb+srv://<user>:<password>@<cluster>.mongodb.net/bettermind
JWT_SECRET=<any-long-random-string>
# Required for sensitive data encryption
ENCRYPTION_SECRET=<any-long-random-string>
ENCRYPTION_SALT=<any-short-string>
# Optional — features degrade gracefully if omitted
OPENAI_API_KEY=sk-... # AI chat (falls back to keyword matching)
GOOGLE_CLIENT_ID=... # Google Calendar sync
GOOGLE_CLIENT_SECRET=...
GOOGLE_REDIRECT_URI=http://localhost:3000/api/auth/google/callbackBetterMind/
├── public/ # Static assets
├── src/
│ └── app/
│ ├── api/ # Route Handlers (auth, mood, assessment, chat, doctors, admin…)
│ │ └── auth/2fa/ # TOTP setup / verify / disable
│ ├── components/ # Navbar, Footer, GuestBanner, GuestGateModal
│ ├── contexts/ # AuthContext (auth + guest mode + auth-gate)
│ ├── dashboard/ # Dashboard + chart components
│ ├── doctor/ # Doctor portal (dashboard, profile, verification)
│ ├── admin/ # Admin verification console
│ ├── lib/ # db, authServer, twoFactorAuth, openaiClient,
│ │ # googleCalendar, slotGenerator, rateLimit, encryption, demoData
│ ├── mood/ assessment/ chat/ doctors/ appointments/ resources/ settings/
│ ├── globals.css # Light design system (tokens, components)
│ └── layout.js
├── tailwind.config.js # Extended theme (brand/mint/calm/ink scales, shadows)
└── README.md
📚 Deep technical documentation lives in the Wiki — architecture, full API reference, auth & security internals, the design system, and guest-mode internals.
Copy-paste-ready bullets describing what this project demonstrates.
- Built a full-stack mental health platform with Next.js 14 (App Router), React 18, MongoDB, and Tailwind CSS, supporting three role-based experiences (patient, doctor, admin).
- Engineered a custom JWT authentication system with httpOnly cookies, bcrypt hashing, TOTP two-factor authentication (QR provisioning + recovery codes via
otplib), and server-side verification with caching. - Integrated OpenAI
gpt-4o-miniinto an AI mental-health companion, injecting the user's recent mood and clinical-assessment data as context for personalized, safety-aware responses, with a graceful keyword-based fallback. - Implemented clinically-validated PHQ-9 & GAD-7 assessments with automated severity scoring, historical trend visualization, and a question-by-question breakdown.
- Designed an interactive analytics dashboard with Recharts — mood trends, assessment trends, activity-impact analysis, and a Pearson correlation between mood and symptom severity.
- Built doctor discovery & appointment booking with availability slots, Google Calendar OAuth sync, ratings/reviews, and an admin verification workflow.
- Added API rate limiting, AES encryption for sensitive data, and role-based access control across the stack.
- Led a complete UI/UX redesign from a dark theme to a research-backed calming light design system (color psychology for mental health), built on Tailwind design tokens and Framer Motion micro-interactions.
- Shipped a “Guest Mode” that lets anyone explore the full product with realistic demo data while keeping all writes safely behind auth — purpose-built for frictionless demos.
- Real-time doctor↔patient messaging (WebSockets)
- Push/email reminders for appointments and check-ins
- Exportable PDF wellness reports
- Mobile app (React Native) sharing the same API
- Expanded assessment library (PSS, WHO-5)
- Fork the repo
- Create a feature branch (
git checkout -b feature/amazing-feature) - Commit (
git commit -m 'Add amazing feature') - Push (
git push origin feature/amazing-feature) - Open a Pull Request
Licensed under the MIT License — see LICENSE for details.
Built with care for mental wellness. 🧠💙