Landing Page
Character Setup
Game Screen

📖 Lorebound

Lorebound Turn your study material into an after class adventure. Your knowledge is your weapon.

👤 Team Information

Field	Details
Name	Naif Ali Bin Dair
Country	Saudi Arabia
Track	Productivity

Project Description

Overview

Lorebound is an AI-powered web application that transforms students' study material into an interactive role-playing game experience. Instead of passively re-reading notes or grinding through flashcards after class, students upload their study PDF, name a character, choose a world, and are immediately dropped into a living narrative — where every question they answer correctly advances their story, and every wrong answer costs them a life and earns them an explanation.

The application is designed for after-class, single-session studying — a focused 15–30 minute experience that reinforces what was just learned through active recall wrapped in engaging storytelling.

The Problem

After-class studying is one of the most important and most neglected stages of learning. Students know they should review their material, but conventional methods — re-reading notes, Quizlet sets, or static practice tests — offer no motivation, no engagement, and no consequence for wrong answers. The result is passive skimming rather than genuine recall.

The gap Lorebound fills is simple: students don't avoid studying because the material is hard — they avoid it because it's boring.

The Solution

Lorebound solves this by combining three proven learning techniques with generative AI:

1. Active Recall — Instead of reading, students are challenged with questions directly generated from their own material, forcing genuine memory retrieval.

2. Consequence-Based Learning — The Hearts system means wrong answers have a cost. Stakes — even fictional ones — increase engagement and focus.

3. Immediate Corrective Feedback — When a student answers incorrectly, the AI doesn't just mark it wrong and move on. It pauses the story, explains the concept clearly, and then continues. The mistake becomes a teaching moment embedded in the narrative.

These three elements are wrapped inside a Gemini-powered generative narrative that is unique to every student's character name, chosen world, and study material — making every session feel personal and immersive.

How It Works

Student uploads PDF + sets character name & world
        ↓
Gemini reads the PDF and builds a narrative world from the content
        ↓
Story begins — the character enters a world themed around the student's input
        ↓
At key story beats, the AI generates a multiple-choice question from the PDF
        ↓
  Correct answer → story advances, hearts maintained
  Wrong answer   → heart lost + AI explains the concept → story continues
        ↓
Session ends when hearts reach 0 or the material is covered

Core Features

🎭 Generative Narrator

Google Gemini acts as a live storyteller. It reads the uploaded PDF and generates a unique narrative world based on the student's character name and world description. No two sessions produce the same story — the AI adapts the setting, tone, and characters to match the student's input.

📄 PDF-to-Quest Integration

The student's study material is the source of truth. Gemini extracts key concepts, definitions, and facts from the uploaded PDF and converts them into multiple-choice challenges that appear organically within the story. There are no pre-written question banks — every question is generated live from the student's actual material.

❤️ Hearts & Consequence System

Students begin with a set number of hearts — their health within the story. Correct answers keep the narrative moving forward. Wrong answers cost a heart and trigger a narrative setback (the character faces a challenge, loses ground, or encounters an obstacle). When all hearts are lost, the adventure ends. This system creates low-stakes but real consequences that keep students alert and engaged.

💬 Real-Time Corrective Explanations

This is the most educationally significant feature. When a student answers incorrectly, the AI does not simply reveal the right answer. It pauses the story and delivers a clear, contextual explanation of why the correct answer is right and what the student misunderstood — using language grounded in the student's own study material. Learning happens at the point of failure.

⚡ Streaming Responses

Gemini's responses are delivered word-by-word in real time, giving the session the feeling of a live storytelling experience rather than a static quiz. This pacing is intentional — it maintains immersion and gives students time to absorb the narrative before each challenge.

Technologies Used

Layer	Technology
AI / LLM	Google Gemini API
PDF Processing	Gemini Document Understanding (native PDF input)
Streaming	Gemini streaming API
Frontend	HTML, CSS, JavaScript, React, Vite
Hosting	Netlify
Language	Arabic / English bilingual support

Hackathon Theme Alignment

Primary Theme: Education Technology

Lorebound directly addresses one of the most persistent challenges in education — student engagement during self-directed study. It does not replace the classroom or the teacher. It specifically targets the after-class window, the period where retention is most at risk and motivation is lowest.

The project demonstrates:

A meaningful, real-world problem with a clear target user (students at any level)
An original application of generative AI beyond chatbot use cases
A functional, working prototype with end-to-end flow
Accessibility considerations (bilingual support, simple onboarding)
Scalability potential across subjects, languages, and education levels

Impact

Metric	Impact
Target Users	Students at secondary and university level
Use Case	After-class review sessions (15–30 min)
Subject Coverage	Any subject with a PDF — fully universal
Language Reach	English and Arabic
Learning Outcome	Active recall, corrective feedback, engagement

Every student, regardless of subject or level, has study material in PDF format. Study RPG requires no integration with institutional systems, no teacher setup, and no prior configuration. A student can go from uploading a PDF to answering questions inside their personalized story in under two minutes.

Future Roadmap

While the current version is scoped as a single-session, after-class tool, the architecture supports meaningful expansion:

Session summary — A post-game report showing which questions were missed and which concepts need review
Voice narration — Text-to-speech for the story segments to deepen immersion
Multiplayer study mode — Competing with classmates using the same PDF
LMS integration — Connecting with platforms like Google Classroom or Moodle for direct material import

Screenshots

See attached UI previews

Screen	Description
Landing Page	Introduction to Study RPG with feature overview and Start Adventure CTA
Character Setup	Form to enter character name, world description, hearts count, and PDF upload
Game Screen	Live narrative with character stats, story text, and A/B/C/D answer buttons