Re: Automated reviews by AI (was Re: [PATCH 0/5] Duplicate entry hardening)

[Junio C Hamano <gitster@pobox.com>]

Christian Couder <christian.couder@gmail.com> writes:

> On Tue, Jun 2, 2026 at 8:16 AM Patrick Steinhardt <ps@pks.im> wrote:
>>
>> Overall I agree that everyone who is a core contributor should also make
>> reviews part of their regular worflow. At least for corporate
>> contributors that might also make it easier to communicate this to their
>> respective employers. Regardless of that, my expectation is that there
>> will be times where it works well, and other times where it works less
>> well.
>
> Sashiko (https://github.com/sashiko-dev/sashiko) is used these days by
> Linux kernel developers and seems to work well for them.
>
> At GitLab and probably in other companies, some of us also use AI to
> review our work before sending it to the mailing list. And yeah, it
> helps find issues before our patches reach the mailing list.
>
> In the same way as we require that patches must pass CI, do we want to
> require that patches "pass" an AI review before they get accepted?

I do not think so.  You (figuratively, not limited to Christian
Couder) are welcome to use whatever tool available to you to help
you polish your submission, and the higher quality your patches are
(e.g., fewer typos and jumps in logic flow that interferes the
thought process of human reviewers), the more helpful you are being
to the community.  The use of GitHub PR initiated CI run falls into
the same category, I think, in that we do not require you to have an
account and trigger the CI there, but you are doing a good service
if you made sure you caught breakages on macOS you do not have
access to otherwise before sending your patches to the list.

But I do not think we should require you to bring your own token
budget to be able to contribute.

> The benefit would be that it would hopefully catch a lot of trivial
> things like indentation, typos/grammos, etc, and a lot of things a bit
> more difficult to spot like memory issues. Perhaps with some amount of
> prompting/configuration (for example pointing it at our
> CodingGuidelines and SubmittingPatches) it could also catch issues
> like style issues, commits that do too many things, refactoring
> opportunities, etc.

Yes.

Similarly, you are welcome to use tools including AI tools to help
you review others' patches, or help sanity check your reviews of
others' patches before you send them out.  The reason why such an
effort is valuable to the community is the same.

But I personally consider that the use of the tools (not limited to
AI tools) is up to each developer.  What counts a lot more is the
quality of the output.  Just like PR driven CI at GitHub is offered
to everybody who wants to participate and is willing to have an
account there, it may help those aspiring developers if automated
review services are made easily available, but it is a different
story to _require_ use of such service.