How to deal with fully LLM-generated merge requests

[David Edmundson]

On Thu, Sep 10, 2026 at 9:29 AM Akseli Lahtinen <akselmo at akselmo.dev> wrote:
>
> On Thursday 10 September 2026 10:27:00 Eastern European Summer Time Vlad Zahorodnii wrote:
> > Hi,
> >
> > We started seeing an increase in number of merge requests that have
> > obvious signs of being generated fully by LLMs. For example, walls of
> > text in merge description, or replying with a wall of text to a review
> > comment in less than a minute, etc.
> >
> > It is easy generating such merge requests, but going through them takes
> > time for us, humans. It's possible to encounter a denial of service type
> > of issue if the number of such MRs continues increasing.
> >

> What I propose is to have something like jellyfin has as the baseline,
> but individual projects can ban them outright if they feel like doing so.

We don't want project specific rules. I've seen it before. It ends up
with people having weird power trips over pointless random things.
Coming up with a new commit message style or different way of doing
braces and it makes everything unpleasant for everyone.

If anyone wants to do that, they can find some hosting elsewhere.

David