How to deal with fully LLM-generated merge requests

[Mark Penner]

Sep 10, 2026 03:07:04 Lasath Fernando <kde at lasath.org>:

> I personally am a big fan of Jellyfin's AI/LLM use policy:
> https://jellyfin.org/docs/general/contributing/llm-policies/

Isn't most of that already covered (in spirit at least) by 
https://community.kde.org/Policies/Commit_Policy ? Could we add some more 
specific wording there about how LLM use can violate the Commit Policy?

One thing in the Jellyfin policy that I'm not sure if we have is the 
prohibition on LLM output in direct communication.

Sep 10, 2026 02:28:11 Vlad Zahorodnii <vlad.zahorodnii at kde.org>:

> I suggest that we update our guidelines to say that it is okay to close 
> merge requests that have clear signs of being fully generated by LLMs 
> right away.


I support that.

I'm not sure about the canary idea. I would be disappointed to see 
`AGENTS.md` files in KDE repos, but if it exposed slop faster maybe it 
would be worth it.