Skip to content

DEPLOYMENT.md

Latest commit

 

History

History
140 lines (106 loc) · 3.1 KB

DEPLOYMENT.md

File metadata and controls

140 lines (106 loc) · 3.1 KB
Raw

Deployment Guide for Nitra Mitra Student Assistant

Overview

The student assistant now supports two modes:

  1. Server Mode (Recommended): Uses a Flask backend hosted on Vercel
  2. Client Mode: Users provide their own API keys (original implementation)

Backend Deployment on Vercel

Prerequisites

  • Node.js installed
  • Vercel CLI (npm i -g vercel)
  • Google AI Studio API key

Steps

  1. Deploy to Vercel:
# In your project root
vercel
  1. Configure Environment Variable:
vercel env add GEMINI_API_KEY
# Enter your Gemini API key when prompted
  1. Deploy Production:
vercel --prod
  1. Update Frontend Configuration:
    • The frontend auto-detects the backend URL
    • For custom URLs, update detectBackendUrl() in assistant.js

Environment Variables

Set these in Vercel dashboard or CLI:

  • GEMINI_API_KEY: Your Google AI Studio API key

Frontend Features

Auto-Detection

  • Automatically detects if backend is available
  • Falls back to client mode if backend is unreachable
  • Shows real-time backend status

Mode Toggle

Users can switch between:

  • Server Mode: Uses hosted backend (no API key needed)
  • Client Mode: Uses personal API key (stored locally)

Status Indicators

  • ✅ Backend ready and configured
  • ⚠️ Backend available but not configured
  • ❌ Backend unavailable
  • 🔄 Checking status...

Local Development

Backend

cd backend
pip install -r requirements.txt
export GEMINI_API_KEY="your_key_here"
python app.py

Frontend

  • Open assistant.html in browser
  • Backend URL auto-detects to http://localhost:5000

Configuration

Backend URL Detection

The frontend automatically detects the backend URL:

  • localhost: http://localhost:5000
  • Vercel domains: Uses same origin
  • Custom: Update detectBackendUrl() method

User Preferences

  • Mode selection is saved in localStorage
  • API keys are stored locally (never sent to backend)
  • Backend status is checked on page load

Troubleshooting

Backend Issues

  1. Check Vercel deployment logs
  2. Verify environment variables are set
  3. Test API endpoints directly

Frontend Issues

  1. Check browser console for errors
  2. Verify network connectivity
  3. Try switching between modes

API Rate Limits

  • Backend mode: Shared quota (managed by admin)
  • Client mode: Individual user quotas

Security

Backend

  • API keys stored as environment variables
  • CORS configured for frontend domain
  • Input validation and error handling
  • No sensitive data logged

Frontend

  • API keys stored locally only
  • Secure HTTPS connections
  • XSS protection via DOMPurify

Monitoring

Backend Health

  • GET / - Health check
  • GET /api/status - Configuration status
  • Monitor Vercel function logs

Frontend Analytics

  • Google Analytics integrated
  • Error tracking in console
  • User interaction metrics

Scaling

Backend

  • Vercel auto-scales serverless functions
  • Monitor usage and adjust quotas
  • Consider caching for frequent queries

Frontend

  • CDN delivery via GitHub Pages
  • Optimized asset loading
  • Responsive design for all devices