YouTube Monitor Bot

Intelligent YouTube channel monitoring with AI-powered summaries delivered via Telegram. Never miss important content from your favorite creators.

Features

• Multi-Channel Monitoring - Track unlimited YouTube channels simultaneously using RSS feeds (no API quota limits)
• AI-Powered Summaries - Automatic video summarization using Google's Gemini 2.0 Flash
• Telegram Integration - Get notifications with relevance scores, key takeaways, and watch recommendations
• Smart Retry Logic - Attempts to fetch transcripts for up to 48 hours for videos that don't have them immediately
• Interest-Based Filtering - Customizable interest profile for personalized relevance scoring
• Skip Shorts - Automatically filters out YouTube Shorts and videos under 5 minutes
• Interactive Bot Commands - Manage channels, check health, and view status via Telegram
• State Tracking - Tracks video processing status and retry attempts

How It Works

1. Hourly Monitoring: Checks all configured YouTube channels for new videos via RSS feeds
2. Transcript Fetching: Attempts to download video transcripts (retries every hour for up to 48 hours if unavailable)
3. AI Analysis: Generates structured summaries using Gemini 2.0 with your interest profile
4. Smart Notifications: Only sends Telegram notifications when summaries are successfully generated
5. Interactive Management: Use Telegram bot commands to add/remove channels and check system health

Quick Start

Prerequisites

• Python 3.8+
• Google Gemini API key (Get one here)
• Telegram Bot token (Create via @BotFather)

Installation

1. Clone the repository

   git clone https://github.com/yourusername/youtube-monitor-bot.git
   cd youtube-monitor-bot

2. Install dependencies

   pip install -r requirements.txt

3. Configure environment variables

   cp .env.example .env
   # Edit .env with your actual API keys

4. Set up your interests (optional)

   Edit config/my_interests.txt to customize the AI's relevance scoring for your preferences.

5. Add YouTube channels

   Edit config/channels.txt and add channels (one per line):

   @GregIsenberg
   https://www.youtube.com/@MyFirstMillionPod
   

Usage

Using the Wrapper Script (Recommended)

# Check for new videos
./monitor.sh check

# Start interactive Telegram bot
./monitor.sh bot

# Check system health
./monitor.sh health

# Test Telegram notifications
./monitor.sh test

Direct Python Commands

Run a one-time check for new videos:

python3 src/youtube_monitor_telegram.py --check

Start the interactive bot:

python3 src/telegram_bot_commands.py --run

Automated Monitoring (Cron)

Set up hourly checks by adding to your crontab:

crontab -e

Add this line (adjust paths as needed):

0 * * * * cd /path/to/youtube-monitor-bot && ./monitor.sh check >> data/logs/monitor.log 2>&1

Or run the bot in the background:

nohup ./monitor.sh bot >> data/logs/bot-commands.log 2>&1 &

Telegram Bot Commands

Send these commands to your bot on Telegram:

Command	Description
/list	Show all monitored channels
/add @channel	Add a new channel to monitor
/remove @channel	Stop monitoring a channel
/health	Check system status and cron job health
/help	Show help message

Examples:

/add @GregIsenberg
/add https://youtube.com/@MyFirstMillionPod
/remove @GregIsenberg

Standalone Tools

YouTube Transcript Downloader

Download transcripts from any YouTube video without API keys:

# Basic usage (readable format)
python3 src/youtube_transcript.py "https://www.youtube.com/watch?v=VIDEO_ID"

# Plain text format
python3 src/youtube_transcript.py "VIDEO_ID" --format plain

# Save to file
python3 src/youtube_transcript.py "VIDEO_ID" --output data/transcripts/video.txt

# Specify language preference
python3 src/youtube_transcript.py "VIDEO_ID" --languages en de fr

Available formats:

• readable (default) - Clean, paragraph-formatted text
• plain - Single continuous text
• timestamped - Includes timestamps for each line
• paragraphs - Auto-grouped into paragraphs
• json - Structured JSON with timestamps

Health Check

Check the monitoring system's health status:

./monitor.sh health
# or
python3 src/healthcheck.py

Configuration Files

File	Purpose
.env	API keys and tokens (create from .env.example)
config/channels.txt	List of YouTube channels to monitor
config/my_interests.txt	Your interest profile for AI relevance scoring
data/state.json	Runtime state (auto-generated)
data/video_state.json	Video retry tracking (auto-generated)

Project Structure

youtube-monitor-bot/
├── src/                          # Source code
│   ├── youtube_monitor_telegram.py   # Main monitoring script
│   ├── telegram_bot_commands.py      # Interactive Telegram bot
│   ├── telegram_notifier.py          # Notification formatting & sending
│   ├── youtube_transcript.py         # Standalone transcript downloader
│   └── healthcheck.py                # System health checker
├── config/                       # User configuration
│   ├── channels.txt                  # Channels to monitor
│   └── my_interests.txt              # Your interest profile
├── data/                         # Runtime data (gitignored)
│   ├── logs/                         # Log files
│   ├── transcripts/                  # Downloaded transcripts
│   ├── summaries/                    # Generated AI summaries
│   ├── state.json                    # Channel state
│   └── video_state.json              # Video retry tracking
├── monitor.sh                    # Convenience wrapper script
├── .env.example                  # Environment template
├── requirements.txt              # Python dependencies
├── README.md                     # This file
└── LICENSE                       # MIT License

Example Notification

🎥 New video from Greg Isenberg

Building a $10M Side Project in Public

🔗 Watch on YouTube

Relevance: 9/10 ⭐⭐⭐⭐⭐

TL;DR:
Greg breaks down his strategy for building profitable side projects
by solving real problems and sharing the journey publicly.

Worth watching?
✅ YES - Packed with actionable advice and real revenue numbers

Key takeaways:
1. Started at $2K/mo MRR, now at $83K/mo in 18 months
2. Distribution strategy: Twitter threads + weekly updates
3. Used Supabase + Next.js stack to ship in 2 weeks

How AI Summaries Work

The bot uses your my_interests.txt profile to:

1. Score relevance (1-10) based on how well the video matches your interests
2. Generate TL;DR - One sentence summary (max 20 words)
3. Explain why you'd care - 2-3 bullet points of relevance
4. Extract key takeaways - Specific details, tools, numbers mentioned
5. Recommend watching - YES/NO/MAYBE with reasoning

Troubleshooting

Bot not receiving videos:

• Check cron job is running: ps aux | grep youtube_monitor
• View logs: tail -f data/logs/monitor.log
• Run manual check: ./monitor.sh check

Transcripts failing:

• Some videos don't have transcripts immediately after upload
• Bot will retry every hour for up to 48 hours
• Check data/video_state.json to see retry counts

Telegram notifications not working:

• Verify .env credentials are correct
• Test notifications: ./monitor.sh test
• Check bot token with @BotFather

Health check issues:

./monitor.sh health

Environment Variables

Create a .env file with these variables:

# Required for AI summaries
GEMINI_API_KEY=your_gemini_api_key_here

# Required for Telegram notifications
TELEGRAM_BOT_TOKEN=your_telegram_bot_token_here
TELEGRAM_CHAT_ID=your_telegram_chat_id_here

Getting API Keys

Gemini API Key:

1. Visit Google AI Studio
2. Create a new API key
3. Copy to .env

Telegram Bot:

1. Message @BotFather on Telegram
2. Send /newbot and follow prompts
3. Copy the bot token to .env
4. Get your chat ID by messaging @userinfobot

Advanced Configuration

Customizing Retry Logic

Edit src/youtube_monitor_telegram.py:39:

MAX_TRANSCRIPT_RETRIES = 48  # Number of hourly attempts

Changing Summary Model

Edit src/youtube_monitor_telegram.py:229:

model = genai.GenerativeModel('gemini-2.0-flash-exp')

Filtering Video Length

Modify the short video threshold in src/youtube_monitor_telegram.py:174:

return seconds < 300  # 300 seconds = 5 minutes

Contributing

Contributions are welcome! Please:

1. Fork the repository
2. Create a feature branch: git checkout -b feature-name
3. Commit changes: git commit -am 'Add feature'
4. Push to branch: git push origin feature-name
5. Submit a pull request

License

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

Acknowledgments

• Built with youtube-transcript-api
• Powered by Google Gemini AI
• Telegram Bot API

Support

• Open an issue for bug reports or feature requests
• Star the repo if you find it useful

---

Made for staying on top of YouTube content without the noise.