Telegram AI Bot

A production-grade Telegram AI chatbot powered by NVIDIA AI APIs with enterprise architecture, MCP integration, and automated CI/CD.

Features

• AI Chat: Natural conversations powered by NVIDIA AI (Llama, Mistral, Mixtral models)
• Context Memory: Maintains conversation history across sessions
• Multi-Model: Switch between different AI models
• Commands: /start, /help, /reset, /ping, /stats, /about, /settings, /admin
• Inline Keyboards: Interactive settings panels
• Webhook/Polling: Supports both modes
• Rate Limiting: Built-in abuse prevention
• Session Management: Automatic cleanup of expired sessions
• Health Endpoint: Production monitoring ready
• Prometheus Metrics: Observability built-in
• Structured Logging: Pino-based with file rotation
• Docker: Containerized deployment
• CI/CD: GitHub Actions pipeline
• Render: One-click deployment

Quick Start

Prerequisites

• Node.js 20+
• Telegram Bot Token (from @BotFather)
• NVIDIA AI API Key (from NVIDIA AI)

Local Development

# Clone and install
git clone <repo-url>
cd telegram-ai-bot
npm install

# Set up environment
cp .env.example .env
# Edit .env with your tokens

# Run in development
npm run dev

# Build
npm run build

# Run production
npm start

Environment Variables

Variable	Required	Description
TELEGRAM_BOT_TOKEN	Yes	Bot token from @BotFather
NVIDIA_API_KEY	Yes	NVIDIA AI API key
NVIDIA_MODEL	No	Default model (default: meta/llama-3.1-8b-instruct)
PORT	No	Server port (default: 3000)
HOST	No	Server host (default: 0.0.0.0)
WEBHOOK_URL	No	Webhook URL (set for production)
WEBHOOK_SECRET	No	Webhook secret token
ADMIN_IDS	No	Comma-separated admin Telegram IDs
LOG_LEVEL	No	Logging level (default: info)

Architecture

src/
├── index.ts           # Application entry point & bot setup
├── config/            # Environment configuration & validation
├── constants/         # App constants, commands, error messages
├── types/             # TypeScript type definitions
├── interfaces/        # Interface contracts
├── database/          # SQLite database layer
├── ai/                # NVIDIA AI service integration
├── services/          # Session management & business logic
├── handlers/          # Command & message handlers
├── middleware/        # Request processing middleware
├── server/            # Fastify HTTP server (health, metrics)
├── mcp/               # MCP tool registration & management
├── browser/           # Browser automation interface
└── utils/             # Logger & utilities

API Endpoints

• GET / - Bot information
• GET /health - Health check (used by Render)
• GET /metrics - Prometheus metrics
• POST /webhook - Telegram webhook receiver

Deployment

See RENDER_DEPLOYMENT.md for detailed deployment instructions.

Quick Deploy

1. Push code to GitHub
2. On Render, create a new Web Service
3. Select Docker runtime
4. Add environment variables
5. Deploy

Testing

npm test           # Run tests
npm run test:watch # Watch mode
npm run test:coverage # Coverage report

License

MIT