CSET RAG Chatbot - AI-Powered Assistant

An intelligent chatbot for Thaddeus Stevens College of Technology's Computer Software Engineering Technology (CSET) program, powered by Google Gemini AI with RAG (Retrieval Augmented Generation).

Features

• AI-Powered Responses - Utilizes Google Gemini 2.5 Flash model
• Smart Context Retrieval - RAG with Pinecone vector database
• Modern UI - Responsive React interface with dark mode support
• Secure Backend - Express.js API with environment-based configuration
• Fast & Lightweight - Built with Vite and TypeScript

Architecture

This application consists of two main parts:

1. Frontend (React + TypeScript + Vite)

   • Modern, responsive UI
   • Real-time chat interface
   • API client for backend communication

2. Backend (Node.js + Express + TypeScript)

   • RESTful API endpoints
   • Google Gemini AI integration
   • Pinecone vector database (with mock support)
   • CORS and security middleware

Quick Start

Prerequisites

• Node.js 18 or higher
• npm or yarn
• Google Gemini API key (Get one here)

Setup

For detailed setup instructions, see SETUP.md

Quick setup:

1. Backend Setup

   cd backend
   npm install
   cp env.example.txt .env
   # Edit .env and add your GEMINI_API_KEY
   npm run dev

2. Frontend Setup (in a new terminal)

   npm install
   npm run dev

3. Access the app

   • Frontend: http://localhost:5173
   • Backend API: http://localhost:3001/api

Project Structure

CSET-chatbot/
├── backend/                 # Express.js backend
│   ├── src/
│   │   ├── config/         # Environment configuration
│   │   ├── constants/      # System prompts & knowledge base
│   │   ├── controllers/    # Request handlers
│   │   ├── middleware/     # Express middleware
│   │   ├── routes/         # API routes
│   │   ├── services/       # Gemini & Pinecone services
│   │   ├── types/          # TypeScript types
│   │   └── server.ts       # Main server file
│   ├── package.json
│   └── README.md
├── components/             # React components
│   ├── ChatBubble.tsx
│   ├── Header.tsx
│   ├── IconComponents.tsx
│   └── MessageInput.tsx
├── services/              # Frontend services
│   ├── apiService.ts     # Backend API client (current)
│   ├── geminiService.ts  # Legacy direct integration
│   └── pineconeService.ts # Legacy mock service
├── App.tsx               # Main React component
├── types.ts              # TypeScript types
├── constants.ts          # Frontend constants
├── package.json
├── SETUP.md             # Detailed setup guide
└── README.md            # This file

Configuration

Backend Environment Variables

Create backend/.env:

PORT=3001
NODE_ENV=development
FRONTEND_URL=http://localhost:5173
GEMINI_API_KEY=your_gemini_api_key_here
USE_MOCK_PINECONE=true

Frontend Environment Variables

Create .env in root:

VITE_API_URL=http://localhost:3001/api

API Documentation

Endpoints

• GET /api/health - Health check
• POST /api/chat/message - Send a message to the chatbot
• POST /api/chat/reset - Reset the chat session

For detailed API documentation, see backend/README.md

Development

Running in Development Mode

Backend (with hot reload):

cd backend
npm run dev

Frontend (with hot reload):

npm run dev

Building for Production

Backend:

cd backend
npm run build
npm start

Frontend:

npm run build
npm run preview

Testing the API

Using curl:

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

# Send a message
curl -X POST http://localhost:3001/api/chat/message \
  -H "Content-Type: application/json" \
  -d '{"message": "What is CSET?"}'

Customization

Knowledge Base

Edit backend/src/constants/systemInstruction.ts to customize:

• System instructions
• Knowledge base content
• AI personality

UI Theme

Modify Tailwind classes in components to customize the look and feel.

Security

• API keys stored securely in backend environment variables
• CORS configured to accept requests only from frontend
• Helmet.js for security headers
• Input validation on all endpoints

Tech Stack

Frontend

• React 19
• TypeScript 5.8
• Vite 6
• Tailwind CSS (via inline styles)

Backend

• Node.js
• Express.js
• TypeScript 5.8
• Google Gemini AI SDK
• Pinecone Vector Database
• Helmet, CORS, Morgan

Troubleshooting

See SETUP.md for common issues and solutions.

License

ISC

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

Support

For issues or questions, please create an issue in the repository.