The AI-Enabled Auto L&D Training Module & Contextual Webpage Assistant is a two-part solution that transforms onboarding, training, and daily productivity at Telus Digital.
-
AI Auto L&D Training & Assessment Generator (https://learnmate-telus.lovable.app/)
Instantly creates structured training modules with quizzes and assessments from any project documentation, validating user skills and cutting onboarding time by up to 70%.
Use this Webhook URL : https://n8n-stage.playment.io/webhook/891bf2bf-1158-465e-bc24-dfed230ef908
-
A context-aware browser assistant that provides real-time, task-specific help directly within tools and webpages, reducing query resolution time by 80%.
Together, these solutions ensure every user is trained, evaluated, and supported on the flyโaccelerating project readiness and improving accuracy.
Onboarding today takes weeks, training is inconsistent, and employees waste time searching for answers or relying on SMEs. Our solution addresses this by:
-
Faster, validated skill onboarding for raters, engineers, and support staff.
-
Real-time guidance without switching contexts.
The result: 130,000+ hours saved annually, higher first-time accuracy, faster project allocation, and scalable workforce enablement across Telus Digital.
AI-LearnMate achieves these goals through three integrated technical components:
- ๐ N8N AI Workflows: Automated AI agents for documentation and tool assistance
- ๐ป Lovable UI: Modern React web interface for AI interactions and workflow management
- ๐ Chrome Extension: Browser-based AI assistant for contextual help on any webpage
AI-LearnMate/
โโโ ๐ n8n_workflows/ # AI workflow automation
โโโ ๐ src/ # Lovable UI (React app)
โโโ ๐ chrome_extension/ # Browser extension
โโโ ๐ public/ # Static assets
โโโ ๐ Configuration files # Package.json, configs, etc.
Before setting up AI-LearnMate, ensure you have:
- Node.js (v18 or higher) - Download here
- Python (v3.8 or higher) - Download here
- Chrome Browser (latest version)
- n8n (for workflow automation) - Installation guide
The n8n workflows provide AI-powered automation for documentation assistance and tool guidance using advanced AI agents.
- AI Tool Guide Agent: Comprehensive tool documentation and assistance
- Supervisely Integration: Specialized workflows for video and point cloud annotation
- Streaming Responses: Real-time AI responses with structured output
- Multi-modal Processing: Handles text, images, and video content
# Install n8n globally
npm install -g n8n
# Or using Docker
docker run -it --rm --name n8n -p 5678:5678 n8nio/n8n# Start n8n
n8n start
# Open browser and go to http://localhost:5678
# Navigate to Workflows > Import from File
# Import the following files:Available Workflows:
n8n_workflows/AI LearnMate Workflow.json- Main AI assistant workflown8n_workflows/Get Video Annotation Info.json- Video annotation helpern8n_workflows/Get Point Cloud Annotation Info.json- Point cloud annotation helper
Set up the following credentials in n8n:
# OpenAI API Configuration
API_KEY=your_openai_api_key
MODEL=gpt-4o-mini # or gemini-2.5-flash
# Webhook Configuration
WEBHOOK_URL=http://localhost:5678/webhook/your-webhook-id- Open each imported workflow
- Configure the webhook URLs
- Set up OpenAI credentials
- Activate the workflows
# Test the main workflow
curl -X POST "http://localhost:5678/webhook/891bf2bf-1158-465e-bc24-dfed230ef908" \
-H "Content-Type: application/json" \
-d '{"prompt": "Explain how to use Supervisely for video annotation"}'A modern React application built with Vite, TypeScript, and shadcn/ui components that provides a web interface for AI interactions.
- Modern React Stack: Vite + TypeScript + React 18
- Beautiful UI: shadcn/ui components with Tailwind CSS
- Form Handling: React Hook Form with Zod validation
- Responsive Design: Mobile-first responsive layout
- AI Integration: Connects to n8n workflows for AI responses
# Navigate to project root
cd AI-LearnMate
# Install dependencies
npm installCreate a .env file in the root directory:
# .env
VITE_N8N_WEBHOOK_URL=http://localhost:5678/webhook/your-webhook-id
VITE_API_BASE_URL=http://localhost:5678# Start development server
npm run dev
# The app will be available at http://localhost:5173# Build the application
npm run build
# Preview production build
npm run preview| Command | Description |
|---|---|
npm run dev |
Start development server |
npm run build |
Build for production |
npm run build:dev |
Build in development mode |
npm run lint |
Run ESLint |
npm run preview |
Preview production build |
- PromptForm: Main input form for AI queries
- HtmlViewer: Displays AI-generated HTML responses
- UI Components: Complete shadcn/ui component library
LearnMate Chrome Extension transforms any webpage into an interactive learning experience with AI-powered assistance.
- ๐ค AI Webpage Assistance: Context-aware help on any website
- ๐ Always-On Guide: Floating chat button accessible everywhere
- ๐ง Smart Content Analysis: Extracts text, code, images, and videos
- โก Streaming Responses: Real-time AI communication
- ๐ง Code Copy Functionality: Enhanced developer experience
- ๐ Universal Compatibility: Works on all websites
# Navigate to chrome extension backend
cd chrome_extension/backend
# Install Python dependencies
pip install -r requirements.txt
# Start the FastAPI backend
python -m uvicorn app:app --host 0.0.0.0 --port 8000 --reloadBackend Dependencies:
fastapi==0.104.1
uvicorn==0.24.0
sentence-transformers==2.2.2
faiss-cpu==1.7.4
numpy==1.24.3
pydantic==2.5.0# 1. Open Chrome and go to chrome://extensions/
# 2. Enable "Developer mode" (toggle in top-right)
# 3. Click "Load unpacked"
# 4. Select the chrome_extension/ folder
# 5. Extension should appear in your toolbarUpdate the backend URL in content_script.js if needed:
const CONFIG = {
backendUrl: 'http://localhost:8000', // Update if different
maxContentLength: 8000,
debounceDelay: 300
};- Visit any website (e.g., https://wikipedia.org)
- Look for the purple ๐ฌ chat button in bottom-right corner
- Click the button to open the chat widget
- Ask a question: "What is this page about?"
- Watch the AI respond with page analysis!
| Endpoint | Method | Description |
|---|---|---|
/chat/stream |
POST | Streaming AI chat responses |
/health |
GET | Backend health check |
/docs |
GET | API documentation |
chrome_extension/
โโโ ๐ manifest.json # Extension configuration
โโโ ๐จ popup.html # Extension popup
โโโ โ๏ธ popup.js # Popup functionality
โโโ ๐ง background.js # Service worker
โโโ ๐ content_script.js # Main content injection
โโโ ๐จ inject_icon.css # Floating button styles
โโโ ๐ icons/ # Extension icons
โโโ ๐ backend/ # Python FastAPI backend
โโโ ๐ app.py # Main API server
โโโ ๐ faiss_index.bin # Vector database
โโโ ๐ docs_meta.json # Document metadata
โโโ ๐ docs/ # Documentation files
# Terminal 1: Start n8n
n8n start
# Terminal 2: Start Lovable UI
cd AI-LearnMate
npm run dev
# Terminal 3: Start Chrome Extension Backend
cd chrome_extension/backend
python -m uvicorn app:app --host 0.0.0.0 --port 8000 --reload
# Terminal 4: Load Chrome Extension
# Follow Chrome extension installation steps aboveThe components can work together:
- Chrome Extension โ Extracts webpage content
- N8N Workflows โ Processes content with AI
- Lovable UI โ Provides web interface for complex queries
Configure the Lovable UI to connect with n8n workflows:
// In your React components
const callN8NWorkflow = async (prompt: string) => {
const response = await fetch(process.env.VITE_N8N_WEBHOOK_URL, {
method: 'POST',
headers: { 'Content-Type': 'application/json' },
body: JSON.stringify({ prompt })
});
return response.json();
};# Test main workflow
curl -X POST "http://localhost:5678/webhook/your-webhook-id" \
-H "Content-Type: application/json" \
-d '{"prompt": "Test query"}'# Run development server and test in browser
npm run dev
# Open http://localhost:5173# Test backend health
curl http://localhost:8000/health
# Expected response: {"status": "healthy", "model_loaded": true}- Open test pages:
chrome_extension/test_manual.html - Test workflows: Import and activate in n8n
- Test UI: Submit prompts in the web interface
Create these files for configuration:
VITE_N8N_WEBHOOK_URL=http://localhost:5678/webhook/your-webhook-id
VITE_API_BASE_URL=http://localhost:5678OPENAI_API_KEY=your_openai_api_key
BACKEND_HOST=0.0.0.0
BACKEND_PORT=8000
MAX_CONTENT_LENGTH=8000| Component | Default Port | Configurable |
|---|---|---|
| n8n | 5678 | Yes (n8n config) |
| Lovable UI | 5173 | Yes (vite.config.ts) |
| Chrome Extension Backend | 8000 | Yes (uvicorn command) |
| Issue | Component | Solution |
|---|---|---|
| Port conflicts | All | Change ports in configuration files |
| Extension not loading | Chrome Extension | Check if enabled in chrome://extensions/ |
| Backend connection failed | Chrome Extension | Ensure FastAPI server is running |
| Workflow not responding | N8N | Check webhook URLs and API keys |
| Build errors | Lovable UI | Run npm install and check Node.js version |
# Check all services
curl http://localhost:5678/health # n8n (if health endpoint exists)
curl http://localhost:5173/ # Lovable UI
curl http://localhost:8000/health # Chrome Extension Backend
# Check processes
lsof -i :5678 # n8n
lsof -i :5173 # Lovable UI
lsof -i :8000 # Chrome Extension Backend# View n8n logs
n8n start --log-level debug
# View Lovable UI logs
npm run dev # Check browser console
# View Chrome Extension logs
# Open DevTools > Console on any webpage
# Check chrome://extensions/ for extension errors
# View Backend logs
python -m uvicorn app:app --log-level debug- โ AI Tool Documentation: Comprehensive tool guides
- โ Streaming Responses: Real-time AI communication
- โ Multi-modal Processing: Text, images, videos
- โ Structured Output: JSON formatted responses
- โ Webhook Integration: REST API endpoints
- โ Modern React Stack: Vite + TypeScript + React 18
- โ Component Library: Complete shadcn/ui components
- โ Form Validation: React Hook Form + Zod
- โ Responsive Design: Mobile-first approach
- โ Theme Support: Dark/light mode ready
- โ Universal Compatibility: Works on all websites
- โ Content Analysis: Smart webpage processing
- โ Code Extraction: Syntax highlighting + copy functionality
- โ Streaming Chat: Real-time AI responses
- โ Non-intrusive UI: Floating button design
- Chrome Extension: See
chrome_extension/TECHNICAL_DOCUMENTATION.mdfor detailed technical docs - N8N Workflows: Check individual workflow JSON files for configuration details
- Lovable UI: Built with Lovable - see original README sections above
- Chrome Extension Backend: http://localhost:8000/docs (FastAPI auto-docs)
- N8N Workflows: Available in n8n interface at http://localhost:5678
- Fork the repository
- Create feature branch:
git checkout -b feature/amazing-feature - Make changes to relevant components
- Test all components work together
- Commit changes:
git commit -m 'feat: add amazing feature' - Push to branch:
git push origin feature/amazing-feature - Open Pull Request
- JavaScript/TypeScript: ES6+, consistent formatting
- Python: PEP 8, type hints, docstrings
- React: Functional components, hooks, TypeScript
- CSS: Tailwind CSS classes, responsive design
This project is licensed under the MIT License - see the LICENSE file for details.
- Rohit Bhandari - Developer
- Ujjawal Kumar Singh - Developer
- Lakshay Sethia - Product Manager
- Priyanka Parida - Product Designer
- Issues: Create GitHub issue with detailed description
- Documentation: Check component-specific documentation
- Discussions: Use GitHub Discussions for questions
When reporting bugs, include:
- Component affected (n8n/Lovable UI/Chrome Extension)
- Steps to reproduce the issue
- Expected vs actual behavior
- Console logs and error messages
- Environment details (OS, browser, versions)