Some of the code (the basic scraper and gpt analysis) is here:
https://news.ycombinator.com/item?id=39539103
My plan is to soon post the whole thing in GitHub and track it there