Hello with the rise of LLM usage I start to get in some open source projects that I maintain PRs from people that clearly vibe code.
Problem with these is that they require a lot of work that I’m not sure pays off. Their descriptions and code comments are overly verbose and take forever to read through. This results in many iterations and reviews for what are supposed to be minor PRs.
Normally I don’t mind putting in the work for PRs in the spirit of mentoring folks. But somehow I’m not sure that applies here?
At the same time i don’t want to stop people just because they vibe as who am I to judge. But perhaps it’s fair to ask certain criteria like: keep your descriptions succinct. Don’t comment the obvious or overly verbose. Don’t add examples for 1+1. Don’t add new dependencies if existing dependencies in the tree already cover what you need.. etc.
Are there known policies like this out there? How to be welcoming while at the same time not get abused by ai slop o.o
