Ash HN: Who is cloning my GitHub repo?

I've seen the same thing. Presumably there's lots of misc crawling and scraping going on so you wouldn't expect it to regular, it's not just one org doint periodic updates.

There are places that archive public repos for preservation reasons - like https://archive.softwareheritage.org/

LLM training models

Why not crawl directly from the github frontend though?

Good question. I'm making the assumption that cloning an entire repo with all the history and configuration is much easier and generates less total request traffic than hitting the front end or even using the API.

Do you have any CI pipelines?