Agreed, huge +1 on adding a cap. This is very low-hanging fruit for
restricting AI spammers from flooding us with PRs.

On Fri, Jul 10, 2026 at 10:07 AM Vincent Beck <[email protected]> wrote:

> +1 on the cap
>
> On 2026/07/10 16:56:27 Jarek Potiuk wrote:
> > Hi all,
> >
> > Quick follow-up on the triage follow-up — one concrete,
> > upstream piece of the puzzle.
> >
> > Alongside triaging PRs after they arrive, it helps to limit how many can
> > pile up in the first place. GitHub has just made API support available
> for
> > a "pull request creation cap": a per-repository limit on how many *open*
> > PRs a user *without write access* can have at one time (committers and
> > collaborators are unaffected). See today's announcement:
> >
> >   https://github.com/community/maintainers/discussions/840
> >
> > To make this usable across ASF projects via .asf.yaml, I opened a PR to
> > infrastructure-asfyaml adding support for it:
> >
> >   https://github.com/apache/infrastructure-asfyaml/pull/111
> >
> > Once it's merged, we'd be able to configure something like:
> >
> >   github:
> >     pull_requests:
> >       creation_cap:
> >         enabled: true
> >         max_open_pull_requests: 5
> >
> > This complements the triage effort rather than replacing it: triage
> decides
> > what happens to PRs once they're open, while the cap keeps any single
> > non-collaborator from opening dozens at once — which is where a lot of
> the
> > low-value / automated inflow comes from.
> >
> > If this direction makes sense to you, I'd appreciate a look and a +1 on
> the
> > Infra PR — the more support it gets, the sooner Infra can merge it and we
> > (and other projects) can turn it on.
> >
> > Best,
> > Jarek
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: [email protected]
> > For additional commands, e-mail: [email protected]
> >
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [email protected]
> For additional commands, e-mail: [email protected]
>
>

Reply via email to