Telegram Chat Summarizer

Telegram bot that collects recent messages from configured group chats, summarizes them with OpenAI, and sends the summary to approved Telegram users.

This repository is a maintained derivative of Georgiy Manuilov's MIT-licensed Telegram chat summarizer. The current version adds environment overrides, rate limiting, chunked summaries, and operational logging.

Features

• Monitors one or more Telegram group chats.
• Summarizes recent text messages with a configurable prompt.
• Sends summaries through a Telegram bot.
• Supports approved summary receivers and chat context switching.
• Splits large histories into chunks before summarization.
• Uses environment variables to override sensitive configuration fields.

Requirements

• Python 3.10+
• Telegram API credentials (api_id, api_hash)
• Telegram bot token
• OpenAI API key

Quick Start

git clone https://github.com/DoctorYanmar/Summarizer.git
cd Summarizer
python -m venv .venv
source .venv/bin/activate
python -m pip install -r requirements.txt
cp config.example.json config.json

Edit config.json, then run:

python app.py config.json

On first Telegram client login, Telethon may ask for a phone/code/password and create a local .session file. Session files are ignored by Git.

Configuration

config.example.json contains placeholders:

{
  "telegram_api_id": 123456,
  "telegram_api_hash": "replace-with-telegram-api-hash",
  "telegram_bot_auth_token": "replace-with-bot-token",
  "bot_username": "replace-with-bot-username",
  "openai_api_key": "replace-with-openai-api-key",
  "chats_to_summarize": [
    {
      "id": "example_group_or_chat_id",
      "summarization_prompt_path": "prompts/example_summarization_prompt.txt"
    }
  ],
  "telegram_summary_receivers": [123456789]
}

Sensitive values can also be supplied through environment variables:

• TELEGRAM_API_ID
• TELEGRAM_API_HASH
• TELEGRAM_BOT_AUTH_TOKEN
• BOT_USERNAME
• OPENAI_API_KEY
• CHATS_TO_SUMMARIZE
• TELEGRAM_SUMMARY_RECEIVERS
• MESSAGE_LIMIT

Do not commit config.json, .env, .session files, or logs.

Project Layout

.
├── app.py                                  # Entrypoint
├── config.example.json                     # Safe configuration template
├── requirements.txt
├── prompts/example_summarization_prompt.txt
└── src/
    ├── main.py                            # Application orchestration
    ├── group_chat_scrapper.py             # Telegram history collection
    ├── summarization.py                   # OpenAI summarization
    ├── envoy_bot.py / bot_core.py         # Telegram bot interface
    ├── bot_handlers.py                    # Commands and callbacks
    ├── config.py                          # Config/env loading
    └── user_management.py                 # Receiver verification/context

Attribution

Original implementation by Georgiy Manuilov, MIT License, with documentation published on Habr: https://habr.com/ru/articles/804111/

License

MIT License. See LICENSE.