Skip to content
BranchesTags

Latest commit

 

History

51 Commits

Folders and files

NameName
Last commit message
Last commit date
backend
backend
 
 
database
database
 
 
frontend
frontend
 
 
ml
ml
 
 
.gitignore
.gitignore
 
 
PROGRESS.md
PROGRESS.md
 
 
README.md
README.md
 
 
SETUP.md
SETUP.md
 
 
package-lock.json
package-lock.json
 
 
package.json
package.json
 
 

Repository files navigation

AI-Powered Learning Platform

Preserve Critical Thinking

An AI-powered learning platform that helps anyone think critically through complex problems using Socratic questioning. Integrated with Canvas LMS and multiple MCP servers for personalized assistance using course materials as RAG context.

🏗️ Project Structure

canvas-mcp-cheatcode/
├── backend/           # Node.js + Express + LangGraph + NVIDIA NIM
│   ├── src/
│   │   ├── api/       # REST API server
│   │   ├── mcp/       # MCP server management
│   │   ├── tools/     # AI tools (Canvas, assignments, etc.)
│   │   └── ...
│   ├── package.json
│   └── README.md
│
├── frontend/          # React + Vite (Lovable generated)
│   ├── src/
│   ├── package.json
│   └── README.md      (To be added after Lovable build)
│
├── package.json       # Root monorepo scripts
└── README.md          # This file

🚀 Quick Start

Prerequisites

  • Node.js 18+
  • npm or yarn
  • NVIDIA API Key (Get one here)
  • Canvas API Token (from your Canvas account)

1. Clone Repository

git clone https://github.com/Brian-Two/nvidia-hacks.git
cd nvidia-hacks

2. Install All Dependencies

npm run install:all

3. Configure Backend

cd backend
cp .env.example .env
# Edit .env and add your API keys:
# - NV_API_KEY (NVIDIA)
# - CANVAS_API_TOKEN (Canvas)

4. Run Development Servers

# From root directory
npm run dev

This runs:

📦 Backend

Technology Stack:

  • Node.js + Express
  • LangGraph (AI workflow)
  • NVIDIA NIM (Llama 3.3 Nemotron)
  • Canvas MCP Integration
  • Multi-MCP Server Support

Features:

  • ✅ Socratic AI tutoring
  • ✅ Canvas LMS integration
  • ✅ Multiple MCP servers (GitHub, Google Drive, Notion, Slack)
  • ✅ RAG-powered assignment help
  • ✅ Streaming chat responses
  • ✅ Course material access

Documentation:

Quick Commands:

cd backend
npm run dev      # Start with auto-reload
npm start        # Production mode
npm run chat     # CLI test mode

🎨 Frontend

Technology Stack:

  • React + TypeScript
  • Vite
  • Tailwind CSS
  • Dark theme with emerald green + purple accents

Features:

  • ✅ Canvas connection onboarding
  • ✅ Assignment dashboard
  • ✅ AI chat workbench
  • ✅ MCP server management
  • ✅ Modern, sleek UI

Setup (After Lovable Build):

cd frontend
npm install
# Create .env file:
echo "VITE_API_URL=http://localhost:3001" > .env
npm run dev

Design System:

  • Background: #0F1419 (very dark)
  • Cards: #1C1F26
  • Primary: #10B981 (emerald green)
  • Accent: #A855F7 (purple)
  • Font: Inter

🔌 MCP Server Support

supports multiple MCP (Model Context Protocol) servers:

Service Type Description
Canvas canvas Learning Management System
GitHub github Code repositories
Google Drive google_drive Cloud storage & docs
Notion notion Notes & knowledge base
Slack slack Team communication
Custom custom Any MCP-compatible server

Users can connect multiple servers and the AI will automatically use them as tools.

📚 Documentation

Backend

Frontend

  • Lovable Prompt - UI build specifications
  • Frontend README (after Lovable build)

🎯 Development Workflow

Running Both Servers

Option 1: Root Scripts (Recommended)

# From nvidia-hacks/ directory
npm run dev              # Run both servers
npm run dev:backend      # Backend only
npm run dev:frontend     # Frontend only
npm run install:all      # Install all dependencies

Option 2: Separate Terminals

# Terminal 1 - Backend
cd backend
npm run dev

# Terminal 2 - Frontend
cd frontend
npm run dev

Making Changes

Backend Changes:

  1. Edit files in backend/src/
  2. Server auto-reloads (using --watch)
  3. Test with curl or Postman

Frontend Changes:

  1. Edit files in frontend/src/
  2. Vite hot-reloads automatically
  3. See changes instantly in browser

🧪 Testing

Backend API

# Health check
curl http://localhost:3001/health

# List assignments
curl http://localhost:3001/api/assignments

# Chat with AI
curl -X POST http://localhost:3001/api/chat \
  -H "Content-Type: application/json" \
  -d '{"message": "Explain entropy"}'

# List MCP servers
curl http://localhost:3001/api/mcp/servers

Frontend

🚢 Deployment

Backend

Environment Variables:

NODE_ENV=production
PORT=3001
NV_API_KEY=your_nvidia_key
CANVAS_API_TOKEN=your_canvas_token
FRONTEND_URL=https://your-frontend-domain.com

Deploy Options:

  • Railway
  • Render
  • Fly.io
  • AWS/GCP/Azure
  • Docker container

Frontend

Environment Variables:

VITE_API_URL=https://your-backend-domain.com

Deploy Options:

  • Vercel
  • Netlify
  • Cloudflare Pages
  • AWS Amplify

🎓 Core Features

1. Socratic Learning

  • AI asks questions instead of giving answers
  • Guides critical thinking
  • Helps understand fundamentals
  • Active recall and elaboration

2. Canvas Integration

  • Direct access to courses and assignments
  • Pull course materials for context
  • Track due dates and priorities
  • Submission status

3. Multi-MCP Support

  • Connect GitHub for code help
  • Use Google Drive for research materials
  • Access Notion notes
  • Reference Slack discussions
  • Custom integrations

4. RAG-Powered

  • Uses actual course materials
  • Grounds responses in what you've learned
  • Cites sources from syllabus/lectures
  • Prevents hallucination

5. Modern UI

  • Dark theme optimized for focus
  • ChatGPT-inspired chat interface
  • Card-based assignment view
  • Responsive design

🛠️ Tech Stack

Backend

  • Runtime: Node.js 18+
  • Framework: Express.js
  • AI: LangGraph + NVIDIA NIM
  • LLM: Llama 3.3 Nemotron 49B
  • MCP: Canvas, GitHub, Google Drive, Notion, Slack
  • Language: JavaScript (ES Modules)

Frontend

  • Framework: React 18
  • Build Tool: Vite
  • Styling: Tailwind CSS
  • Language: TypeScript
  • UI Components: Custom + shadcn/ui

📖 Learning Philosophy

  • Active Recall: Retrieve information from memory
  • Elaboration: Connect new knowledge to existing
  • Interleaving: Mix different topics for deeper learning
  • Spaced Repetition: Review at optimal intervals
  • Metacognition: Think about your thinking
  • Socratic Method: Questions lead to discovery

The AI never gives direct answers. It guides you to discover them yourself.

🤝 Contributing

This project is open source! Contributions welcome.

Areas for Contribution

  • New MCP server integrations
  • Enhanced RAG capabilities
  • Study plan generation
  • Mobile apps
  • Additional AI models
  • UI/UX improvements

Development Setup

  1. Fork the repository
  2. Create a feature branch
  3. Make your changes
  4. Test thoroughly
  5. Submit a pull request

🔒 Security & Privacy

  • API Keys: Stored server-side, never exposed to frontend
  • Canvas Data: Stays local, not sent to third parties
  • User Privacy: No data collection or tracking
  • Open Source: Full transparency of code

📝 License

MIT License - See LICENSE file

🆘 Support

🗺️ Roadmap

Phase 1 (Current) ✅

  • Canvas integration
  • Basic chat interface
  • MCP server management
  • Dark theme UI

Phase 2 (Next)

  • Study plan generation
  • Spaced repetition system
  • Progress tracking
  • Mobile responsiveness

Phase 3 (Future)

  • Vector database RAG
  • Multi-user support
  • Study groups
  • Instructor dashboard

See FUTURE_VISION.md for complete roadmap.

🌟 Credits

Built with:

  • NVIDIA NIM - LLM inference
  • LangGraph - AI workflow orchestration
  • Canvas LMS - Learning management integration
  • Lovable - Frontend development

📞 Contact

Made with ❤️ to bring back critical thinking in learning

About

No description, website, or topics provided.

Resources

Readme
Activity

Stars

2 stars

Watchers

0 watching

Forks

0 forks
Report repository

Releases

Packages

Contributors

Languages