Turn your documents into a full study system — powered by the AI model you already pay for.
Upload a PDF, pick your provider, and get flashcards, structured notes, a study guide, and a gamified learning roadmap in seconds.

---

Demo

▶ Watch demo

---

Screenshots

Dark	Light

Sign-in page

---

Features

• Bring your own AI — Gemini, OpenAI, Claude, OpenRouter. Configure once, swap per feature. We never touch your API keys.
• Flashcards + spaced repetition — cards you struggle with surface more often; mastered cards fade out.
• Study guides — structured notes extracted from your document, section by section.
• Gamified roadmap — a Duolingo-style learning path through your material. Complete topics to unlock the next, earn XP, track progress.
• Background generation — close the tab and come back. Your API key is RSA-encrypted in transit and deleted the moment the job finishes.
• Vision-aware — flagship models read PDFs directly; diagrams, equations, and scanned pages stay intact.
• Sync across devices — projects, flashcards, and study guides synced via your account.
• Auth built in — email/password + Google OAuth, password reset, email verification.
• Tier system — Free, Student, and Pro plans with enforced project and file limits.
• Self-hostable — Docker Compose options for bundled MongoDB or MongoDB Atlas; deploys to Coolify in minutes.
• Registration control — REGISTRATION_MODE=invite-only locks signups; admin creates accounts manually.

---

Stack

Layer	Tech
Frontend	Next.js 16, Tailwind CSS, shadcn/ui
Backend	Express 4, TypeScript, Better Auth
Database	MongoDB 8 (bundled or Atlas)
Email	Resend
Auth	Better Auth (email/pw + Google OAuth)
Payments	Stripe (optional)
Containers	Docker, GHCR

---

Monorepo structure

exam-helper/
├── web/      # Next.js frontend   — ghcr.io/anasyd/exam-helper-web
└── server/   # Express backend    — ghcr.io/anasyd/exam-helper-server

---

Self-hosting

See docs/self-hosting.md for the Coolify deployment guide, Compose file comparison, generated secrets, and smoke-test checklist.

---

Local development

Backend

cd server
cp .env.example .env   # fill in values
npm install
npm run dev            # http://localhost:4000

Frontend

cd web
cp .env.example .env.local   # defaults work for local dev
npm install
npm run dev                  # http://localhost:3000