Reverse engineered prompt

GitHub

Build me a Telegram news aggregator like this repo is aiming for. I want it to pull posts from multiple Telegram channels, group similar news together so duplicates do not flood the feed, and publish a single combined feed to a Telegram channel. It should also separate sources into a few trust or source groups so readers can quickly see what kind of channel the news came from.

Please get the whole thing working end to end in Python, with the database running in Docker, and make the crawler and sender easy to start with simple scripts. I should only need to add my Telegram API credentials and, if useful, an OpenAI API key for any LLM based features. If models need to be downloaded, wire that up too.

Keep it practical and close to the project described here, not a totally new product. Make sensible defaults, include a short setup guide, and make sure I can run it locally and test the news collection, clustering, and posting flow. Look up current docs online if you need to.

Have a live product UI? Try website reverse