🎴 MeshCards

AI-Powered Flashcard Generator - Transform any content into Anki-ready flashcards instantly

Live Demo • Documentation • Report Bug • Request Feature

✨ Features

AI-Powered Generation

Multiple AI Models: Gemini 3 Pro, Gemini 3 Flash, Gemini 2.5 Series (Pro/Flash/Lite)
Latest Tech: Powered by Gemini 2.0 Flash and 1.5 Pro
Smart Extraction: Automatically identifies key concepts and creates Q&A pairs
LaTeX Support: Perfect for math, science, and technical subjects
Multimodal: Generate flashcards from text, PDFs, and images

Content Sources

📝 Text Input: Paste notes, articles, or study materials
📄 PDF Upload: Extract content from textbooks and papers
🖼️ Image Upload: Generate from diagrams, charts, and screenshots
📖 Multiple Files: Combine multiple sources in one deck

Customization

Difficulty Levels: Beginner, Intermediate, Advanced
Card Styles: Q&A, Cloze Deletion, Mixed
Deck Size: 10-100 cards per generation
Custom Instructions: Fine-tune generation with specific requirements

User Features

Authentication: Secure sign-in with Supabase
Daily Quota: 2 free decks per day (resets at 12 AM IST)
Usage Tracking: Monitor your daily generation count
Instant Download: Get .apkg files ready for Anki

Security & Limits

4-Layer Protection: Frontend validation, authentication, quota system, rate limiting
Privacy First: No data stored, instant generation and download
API Protection: Automatic error reporting and abuse prevention
IST Timezone: Fair quota reset for all users

Quick Start

Prerequisites

Python 3.11+
Node.js 18+
Supabase account (for auth & quotas)
Gemini API key (or OpenAI/Anthropic)

Installation

Clone the repository

git clone https://github.com/pranavsinghpatil/MeshCards.git
cd MeshCards

Backend Setup

# Create virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

# Configure environment
cp .env.example .env
# Edit .env with your API keys

Frontend Setup

cd frontend
npm install

# Configure environment
cp .env.example .env
# Edit .env with your API URL

Run Development Servers

# Terminal 1 - Backend
uvicorn backend.main:app --reload --port 8000

# Terminal 2 - Frontend
cd frontend
npm run dev

Open in Browser
```
http://localhost:8080
```

📖 Usage

Basic Workflow

Sign In - Authenticate with Google/Email
Input Content - Paste text or upload files
Configure - Choose model, difficulty, and style
Generate - AI creates flashcards in seconds
Download - Get .apkg file for Anki
Import - Open in Anki and start studying!

Example: Generate from Text

Input: "Photosynthesis is the process by which plants convert light energy into chemical energy..."

Output: 
Q: What is photosynthesis?
A: The process by which plants convert light energy into chemical energy

Q: What are the main inputs of photosynthesis?
A: Light energy, water, and carbon dioxide

Q: What are the products of photosynthesis?
A: Glucose and oxygen

Example: Generate from Image

Upload a diagram → AI analyzes visual content → Creates flashcards about the diagram

Architecture

Tech Stack

Backend:

FastAPI (Python web framework)
Supabase (Authentication & database)
Google Gemini AI (Primary LLM)
Genanki (Anki deck generation)
SlowAPI (Rate limiting)

Frontend:

React 18 + TypeScript
Vite (Build tool)
TailwindCSS (Styling)
Shadcn/ui (Components)
React Router (Navigation)

Deployment:

Backend: Leapcell
Frontend: Vercel
Database: Supabase

Project Structure

MeshCards/
├── backend/
│   ├── core/
│   │   ├── llm.py              # LLM client management
│   │   ├── generator.py        # Flashcard generation logic
│   │   ├── anki.py             # Anki deck builder
│   │   ├── auth.py             # Authentication & quota
│   │   ├── error_reporter.py  # Automatic error reporting
│   │   └── storage.py          # File storage management
│   └── main.py                 # FastAPI application
├── frontend/
│   ├── src/
│   │   ├── components/         # React components
│   │   ├── pages/              # Page components
│   │   ├── hooks/              # Custom hooks
│   │   └── lib/                # Utilities
│   └── public/                 # Static assets
├── docs/                       # Documentation
└── decks/                      # Generated decks (local dev)

🔧 Configuration

Environment Variables

Backend (.env):

# Environment
ENV=development  # or production

# AI Providers
GEMINI_API_KEY=your_gemini_key
OPENAI_API_KEY=your_openai_key  # Optional

# Supabase (Required)
SUPABASE_URL=https://your-project.supabase.co
SUPABASE_ANON_KEY=your_supabase_anon_key
SUPABASE_SERVICE_ROLE_KEY=your_supabase_service_role_key

# Rate Limiting
RATE_LIMIT_FREE=2/hour

# GitHub Integration (Optional)
GITHUB_TOKEN=ghp_your_token
GITHUB_REPO=username/repository

Frontend (.env):

# API Configuration
VITE_API_URL=http://localhost:8000

# Maintenance Mode
VITE_MAINTENANCE_MODE=false

Features in Detail

Quota System

Limit: 2 decks per user per day
Reset: 12:00 AM IST (India Standard Time)
Tracking: Stored in Supabase profiles table
Enforcement: Strict in production, lenient in development

Rate Limiting

Limit: 2 requests per hour per IP address
Purpose: Prevent abuse and bot attacks
Independent: Works even without authentication
Bypass: Difficult (requires IP change)

Error Reporting

Automatic: Creates GitHub issues for backend errors
Privacy: Sanitizes API keys and sensitive data
Context: Includes stack trace, user ID (anonymized), and request data
Labels: Auto-categorized (bug, llm, quota, auth)

Feedback System

File Upload: Attach screenshots and files
GitHub Issues: Automatically creates issues
Categories: Bug, Feature Request, Praise, Other
Rating: 1-5 star rating system

Deployment

Backend (Leapcell)

Create Project on Leapcell
Connect Repository
Set Environment Variables
Configure Build:
- Build Command: pip install -r requirements.txt
- Start Command: uvicorn backend.main:app --host 0.0.0.0 --port 8000
Deploy

Full Guide

Supabase Setup

Create Project on Supabase

Create profiles table:

CREATE TABLE profiles (
    id UUID PRIMARY KEY,
    daily_count INTEGER DEFAULT 0,
    last_reset DATE NOT NULL,
    created_at TIMESTAMP DEFAULT NOW()
);

Enable RLS Policies (Guide)
Get API Keys from Project Settings

Contributing

We welcome contributions! Here's how:

Fork the repository
Create a feature branch (git checkout -b feature/AmazingFeature)
Commit your changes (git commit -m 'Add some AmazingFeature')
Push to the branch (git push origin feature/AmazingFeature)
Open a Pull Request

Development Guidelines

Follow existing code style
Add tests for new features
Update documentation
Keep commits atomic and descriptive

Documentation

Mission & Purpose: The problem we solve.
Quick Start User Guide: Get generating in minutes.
System Architecture: How it works under the hood.
Auth & Quotas: Understanding limits and BYOK.
Production Guide: Full deployment checklist.
Developer Guide: Setup and debugging.

Known Issues

LaTeX rendering may fail for complex equations (use $$ delimiters)
Large PDFs (>10MB) may timeout
Image generation requires specific model selection

See Issues for full list.

Roadmap

Short Term

Add more AI models (Claude, Llama)
Implement deck preview before download
Add deck history and management
Support for more file formats (DOCX, PPTX)

Medium Term

Premium tier with higher quotas
Collaborative deck creation
Deck sharing and marketplace
Mobile app (React Native)

Long Term

Spaced repetition algorithm integration
Study analytics and insights
Community features
Offline mode

Support

If you find MeshCards helpful, consider supporting the project:

Star this repository
Buy me a coffee
Report bugs and suggest features
Share with fellow students

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

Google Gemini - For powerful AI capabilities
Anki - For the amazing spaced repetition system
Supabase - For authentication and database
FastAPI - For the excellent Python framework
React - For the frontend framework
All Contributors - For making this project better

Contact

Pranav Singh Patil

Stats

Daily Active Users: Growing!
Decks Generated: Thousands
Success Rate: 98%+
Average Generation Time: 6-12 seconds

Made with ❤️ for students worldwide

⬆ Back to Top

Name		Name	Last commit message	Last commit date
.github/workflows		.github/workflows
assets		assets
backend		backend
decks		decks
docs		docs
frontend		frontend
supabase		supabase
.env.example		.env.example
.gitignore		.gitignore
CONTRIBUTING.md		CONTRIBUTING.md
Dockerfile		Dockerfile
README.md		README.md
launch.bat		launch.bat
requirements.txt		requirements.txt
vercel.json		vercel.json

Latest commit

History

Folders and files

Repository files navigation

🎴 MeshCards

✨ Features

AI-Powered Generation

Content Sources

Customization

User Features

Security & Limits

Quick Start

Prerequisites

Installation

📖 Usage

Basic Workflow

Example: Generate from Text

Example: Generate from Image

Architecture

Tech Stack

Project Structure

🔧 Configuration

Environment Variables

Features in Detail

Quota System

Rate Limiting

Error Reporting

Feedback System

Deployment

Backend (Leapcell)

Supabase Setup

Contributing

Development Guidelines

Documentation

Known Issues

Roadmap

Short Term

Medium Term

Long Term

Support

License

Acknowledgments

Contact

Stats

About

Resources

Contributing

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages