On 12/10/26 00:06, Sharlatan Hellseher wrote:
As we have added a light version of GCD008 in manual [1] on acceptance of GenAI changes to code base, I've added a related tag "kind/gen-ai-suspect" to mark PRs which looks suspension and have to be checked with an author before starting the review process to save time.
You can just ask them? How is any different from other problems that also do not get their own tag?
As far as I understand - we need to close PR wich for sure is assisted or 100% generated by LLM without any review proces or time spent on explaining the reason [2]?
That is a very strange conclusion. We got to meet people where they are somewhat and try to get them to become a contributor.
Guix has a very extreme stance currently, so we should put some effort into explaining that stance before shunning people.
We will never get the amount of contributors we need to maintain all packages if we close P.R.s without explaining why. Branding P.R.s with such a tag is already pretty extreme.
We are not a cult. (Right?)
Please feel free to remove the tag or change it's name/description, I just needed something to filter out obvious LLM generated code during PR run. [1]. https://codeberg.org/guix/guix/pulls/11407 [2]. https://codeberg.org/guix/guix/pulls?q=&type=all&sort=relevance&labels=2241945&state=open&milestone=0&project=0&assignee=0&poster=0 --- Oleg
