Policy on accepting PRs that include LLM generated code?

[astatide]

Hello!

I've just been checking out Helix recently, and I quite enjoy what I've seen so far! It's fast, intuitive, ships a lot of nice defaults, and works well with what I've used it for thus far!

I apologize if this has been asked before (I was unable to find anything discussing it previously); I was curious if the project had any specific guidelines on whether it would be utilizing LLMs as part of the development process, or whether the project would accept PRs that had involved Claude, CoPilot, ChatGPT, etc, in some fashion?

(My own personal interest in asking is that I am actively disinterested in using software stacks that incorporate the usage of LLMs anymore than I am forced to).

Thank you!

[the-mikedavis]

We don't currently have a policy but there have been more and more PRs and other interaction (comments, issues, etc.) that seem LLM-generated so I think it's getting time to adopt one. @archseer and I were discussing this the other day and we like Ghostty's policy: https://github.com/ghostty-org/ghostty/blob/main/AI_POLICY.md.

I think LLMs are good at certain tasks like building timelines or helping you dig into unfamiliar code. But it's very easy to abuse them to waste peoples' time. And I find the output quality pretty questionable most of the time.

[TudbuT]

I'm a long time helix user (not sure how long exactly, but its been my main and only editor and coding environment for most of the last 2 years for sure) as well. I don't actually care about the functionality or soullessness that much, but I do have the following to say:

1. My websites (https://git.tudbut.de among others) are currently, that is right now, receiving 40+ requests per second, so over 3.4 million requests per day (proof), from LLM scanners to gather training data. This has on several occasions caused databases to corrupt due to disks filling up with caches, garbage, logs, etc. Installing a filter (anubis) did not help, though I eventually made my own filter which now does block them reliably. Those who use and pay for LLMs thereby fund a denial of service attacks against millions of smaller websites and services.
2. LLM use degrades skill and is expensive: The more you offload your coding, the less you code, the less you practice, and thus the worse your "coding senses" get. You dont spot mistakes as easily, you get worse at troubleshooting, I've even heard of cases where someone basically unlearned coding for the most part. LLMs are also very exclusive, meaning you are at the mercy of their hosters and their pricing.
3. LLM output looks really good, often enough. I know because I have tried them (specifically Claude Sonnet 4.5, one of the most recent and highest-rated models). It is easy to get them to write a great looking pull request (or other flavor of commit or set thereof), which seems like a good thing, but usually the result is less great than it looks: LLMs are specifically trained for looks, because that is the way the training process works, but great looks can and do often come with micro-bugs and suboptimal architecture/api decisions that are invisible at first but cause severe codebase degradation over time.
4. They make human contributions harder. - LLM code is much harder to read for humans because a) its not specifically trained to be readable, b) its a mixture of everything, including things that really do not fit the rest of the codebase or that are of low quality, and c) LLMs are already beginning to be fed their own excrement, resulting in the beginnings of model collapse (the self-amplification of LLM-typical patterns, many of which are already hard to read for humans due to the previous points).

[GwenTheKween]

As a long time vim user (10+ years) who just decided to switch to helix because vim (and neovim) started accepting LLM contributions, I would very much like to see helix take a stance against it.
I enjoyed my time using it so far, but to be fully transparent, there's very little chance I'll ever contribute beyond issues and sharing the project, since I don't program in rust, so take my feedback however you want it

[nbsp]

throwing my hat in the ring, even if i don't have much to add: i would be massively disappointed if Helix accepts AI contributions. you should have an explicit policy, but that policy should be that LLM use is not accepted, and that hiding LLM use is inappropriate conduct.

[Efimero]

I have used helix for all of a month, specifically because it was not tainted by LLM shenanigans, so if it's tainted, I'll just stop using it...
I'll go back to EDIT.EXE for MS-DOS if that's what I have to do to get away from LLMs...

[okennedy]

I've been using Helix for the better part of a few months. It's the editor that finally got me to give up emacs keybindings for modal editing, and I haven't looked back. LSP integration makes sense as an extensibility model -- it's as powerful, but not nearly as magical as Sublime or emacs (in a good way).

Others above have already addressed how LLM corps are having negative impacts on the environment and open commons. I'll merely state that I do my best, as an individual to avoid LLM-contaminated work on ethical grounds...

... but there's a far more pragmatic reason to reject them: AI code contributions are not a good sign for the health of a project. Creating maintainable code (that is, maintainable without shelling out for a Claude/Gemini/etc... subscription) with an AI is presently hard. Numerous projects have complained about the maintenance overhead of trying to sort through AI slop pushes. Just cut it off at the source.

There's also this well sourced list of complains.

[wolfskaempf]

As a long time helix user who tries to avoid LLM poisoned code as much as practicable, I would be sad to see anything except a hard no-LLM stance from this wonderful project.

The points I would make to support this have already been made above so I won’t repeat them, but one thing I would like to add: Just as with asbestos there were even many people trying to do the right thing, but didn’t know yet about all the harm it causes. With LLM generated text that looks like code we’ve already found many harms today, the low hanging fruit, the long term harm will only become visible later and I would love to see helix survive long-term.

thanks ❤️