Bots Overview

MMPS includes 6 specialized Telegram bots, each designed for specific purposes. All bots run independently with their own databases, services, and scheduled tasks.

The 6 Bots

1. Chatbot - AI-Powered Assistant

The main conversational bot powered by OpenAI and Anthropic.

• Features: 20+ AI tools, reminders, weather, exercise tracking, GitHub integration
• LLM: ChatGPT or Claude
• Status: Production-ready and actively used
• Learn more →

2. Coach - Sports Analytics & Predictions

Real-time sports data with match analysis and predictions.

• Features: Match summaries, competition tables, betting value analysis
• Data Source: Scores365 API
• Status: Production-ready
• Learn more →

3. Langly - Language Learning Assistant

Interactive language learning companion.

• Features: Daily challenges, vocabulary building, language practice
• Status: Production-ready
• Learn more →

4. Magister - Course Management

Learning management system for course tracking.

• Features: Lesson reminders, progress tracking, course management
• Status: Production-ready
• Learn more →

5. Wolt - Restaurant Notifications

Monitors and notifies about restaurant availability.

• Features: Availability monitoring, push notifications
• Status: Production-ready
• Learn more →

6. Worldly - Geography Education

Geography teaching and trivia challenges.

• Features: Trivia challenges, geography education, location-based content
• Status: Production-ready
• Learn more →

Running Bots

Development Mode (One Bot)

LOCAL_ACTIVE_BOT_ID=chatbot npm run start:dev

Replace chatbot with any of: coach, langly, magister, wolt, worldly

Production Mode (All Bots)

IS_PROD=true npm start

All bots run in parallel with separate databases.

Bot Architecture

Each bot follows the same architecture:

features/{bot}/
├── {bot}.init.ts              # Initialization
├── {bot}.controller.ts        # Telegram handlers
├── {bot}.service.ts           # Business logic
├── {bot}-scheduler.service.ts # Scheduled tasks
├── {bot}.config.ts            # Configuration
└── types.ts                   # Type definitions

Configuration

Each bot has a BOT_CONFIG in its config file:

export const BOT_CONFIG = {
  id: 'CHATBOT',
  name: 'Chatbot',
  token: 'CHATBOT_TELEGRAM_BOT_TOKEN',
  commands: { START, HELP, /* ... */ },
};

Database Isolation

Each bot has its own MongoDB database:

• chatbot-db - Chatbot data
• coach-db - Coach data
• langly-db - Langly data
• magister-db - Magister data
• wolt-db - Wolt data
• worldly-db - Worldly data

Shared Features

AI Tools (Chatbot & Coach)

• Weather information
• Reminders and notifications
• Calendar integration
• GitHub API access
• And 20+ more tools

External Services

All bots can access shared services:

• MongoDB for persistence
• Google Cloud services
• External APIs
• Telegram API (via grammY)

Monitoring

In production, all bots log to Google Sheets for monitoring:

SHEETS_CLIENT_EMAIL=...
SHEETS_PRIVATE_KEY=...
SHEETS_LOGS_SPREADSHEET_ID=...

Environment Variables

Each bot requires its own Telegram bot token:

# Required for all bots
MONGO_URI=mongodb://...

# Bot tokens (add only for bots you want to run)
CHATBOT_TELEGRAM_BOT_TOKEN=...
COACH_TELEGRAM_BOT_TOKEN=...
LANGLY_TELEGRAM_BOT_TOKEN=...
MAGISTER_TELEGRAM_BOT_TOKEN=...
WOLT_TELEGRAM_BOT_TOKEN=...
WORLDLY_TELEGRAM_BOT_TOKEN=...

Next Steps

Select a bot to explore:

• Chatbot - AI assistant with tools
• Coach - Sports analytics
• Langly - Language learning
• Magister - Course management
• Wolt - Restaurant finder
• Worldly - Geography education

Or explore:

• Architecture - System design
• Development - Contributing
• Deployment - Production setup