Lutris maintainer use AI generated code for some time now. The maintainer also removed the co-authorship of Claude, so no one knows which code was generated by AI.
Anyway, I was suspecting that this “issue” might come up so I’ve removed the Claude co-authorship from the commits a few days ago. So good luck figuring out what’s generated and what is not.
I have multiple contributions to FOSS projects (incl. small patches to bash and linux), and I’m a commit-access maintainer of Nixpkgs. While I do use a local LLM occasionally for boilerplate stuff, the attitude here makes me immediately distrustful of the Lutris maintainers. Going out of your way to disable the Claude Co-Authored-By on commits is counterproductive and dangerous to the project in the long run. LLMs can often confidently spit out good-looking code with subtle but critical flaws. Commits with non-trivial amounts of LLM output need way more scrutiny than human-written commits, and making it more difficult to tell the two apart is fucked up.