This project is a fullstack chatbot that can impersonate Xavi Amatriain using Google's Gemini AI. It features a React frontend with a Node.js backend that supports both OpenAI and Google Gemini backends with automatic switching.

• Dual Backend Support: Switch between OpenAI GPT and Google Gemini
• Dynamic Backend Switching: Switch between backends during runtime via the chat interface
• Xavi Amatriain Impersonation: AI responds as if it were Xavi Amatriain
• Knowledge Base Integration: Uses PDF documents about Xavi's work and background
• File API Support: Uploads PDF files to Gemini File API for native document processing
• Conversation History: Maintains context across chat sessions
• Backend Switching: Easy configuration to switch between AI providers

• Built with React Chatbot Kit
• Modern UI with customizable styling
• Real-time chat interface

• Primary: Google Gemini with File API support
• Fallback: OpenAI GPT (optional)
• Backend Switcher: Automatic routing between AI providers
• File Upload: Native PDF processing with Gemini File API

• Node.js (v18 or higher)
• Google Gemini API key
• (Optional) OpenAI API key

1. Clone the repository:

git clone https://github.com/xamat/Xavibot.git
cd Xavibot

2. Install dependencies:

npm install

3. Set up environment variables in .env:

# API Configuration
REACT_APP_API_URL=http://localhost:8080
BACKEND_TYPE=gemini

# Gemini Configuration
GEMINI_API_KEY=your_gemini_api_key_here
GEMINI_MODEL=gemini-2.5-flash

# Server Configuration
PORT=8080
CORS_ORIGIN=*

# Optional: OpenAI Configuration
OPENAI_API_KEY=your_openai_api_key_here

4. Start the backend:

cd src/server && node index.js

5. Start the frontend (in a new terminal):

npm run start-frontend

6. Open http://localhost:8081/Xavibot in your browser

• npm run start-frontend - Launches the React frontend on port 8081
• npm run dev-server - Launches the server using nodemon for development
• npm run dev - Launches both frontend and backend concurrently
• npm run dev-gemini - Launches with Gemini backend
• npm run dev-openai - Launches with OpenAI backend

• npm run build - Builds the React frontend for production
• npm run deploy - Builds and deploys to GitHub Pages

• npm run test-backends - Tests backend switching functionality

The system supports two types of backend switching:

The system automatically switches between backends based on the BACKEND_TYPE environment variable:

• BACKEND_TYPE=gemini - Uses Google Gemini (default)
• BACKEND_TYPE=openai - Uses OpenAI GPT

You can switch between backends during a conversation by typing specific commands:

• Type /useGemini or /useOpenAI to change backends
• The system will automatically create a new thread for the new backend
• Your conversation history will be preserved within each backend

Note: When switching backends, you'll get a new conversation thread, so previous messages won't be available in the new backend. Gemini is used by default.

The chatbot uses PDF documents stored in src/server/:

• xamatriain.pdf - Main knowledge base about Xavi
• xamatriain_guide.pdf - Additional information
• blog.pdf - Blog content

These files are automatically uploaded to Gemini File API on startup.

The system uses comprehensive instructions to make the AI impersonate Xavi Amatriain, including:

• Professional background and expertise
• Personal communication style
• Knowledge limitations and honesty
• Use of provided documents

1. Set REACT_APP_API_URL=http://localhost:8080 in .env
2. Start backend: cd src/server && node index.js
3. Start frontend: npm run start-frontend

See:

• DEPLOYMENT.md
• TESTING.md

• GEMINI_API_KEY - Your Google Gemini API key
• REACT_APP_API_URL - Backend server URL

• The browser initializes through /session/init and receives a server-managed session id.
• Do not expose XAVIBOT_API_KEY or any equivalent secret in frontend build variables.

• BACKEND_TYPE - Set to 'gemini' or 'openai' (default: 'gemini')
• GEMINI_MODEL - Gemini model to use (default: 'gemini-2.5-flash')
• OPENAI_API_KEY - OpenAI API key (if using OpenAI backend)
• PORT - Server port (default: 8080)
• CORS_ORIGIN - CORS origin (default: '*')

Xavibot/
├── src/
│   ├── server/
│   │   ├── gemini-backend.js      # Gemini backend implementation
│   │   ├── openai-backend.js      # OpenAI backend implementation
│   │   ├── backend-switcher.js    # Backend switching logic
│   │   ├── config.js              # Configuration and instructions
│   │   ├── app.js                 # Main Express server
│   │   └── index.js               # Server entry point
│   ├── ActionProvider.js          # Frontend action handling
│   ├── MessageParser.js           # Message parsing logic
│   └── ChatbotContainer.js        # Main chatbot component
├── public/                        # Static assets
├── .env                           # Environment variables
└── package.json                   # Dependencies and scripts

• DEPLOYMENT.md - Backend/frontend deployment steps
• TESTING.md - Post-deploy smoke tests and regression checklist
• GEMINI_MIGRATION.md - Migration guide from OpenAI to Gemini
• BACKEND_SWITCHING.md - Backend switching implementation details
• KNOWLEDGE_BASE_STRATEGIES.md - Knowledge base strategies and approaches

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

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

• Built with React Chatbot Kit
• Powered by Google Gemini and OpenAI
• Knowledge base contains information about Xavi Amatriain's work and background