On Tue, Oct 27, 2026 at 08:58:15PM +0100, Lucas Nussbaum wrote: > I don't see the point in hiding those reports. If some maintainers > prefer to wait until another human confirmed the diagnosis, that's fine, > but why penalize those who welcome early feedback?
It's not about hiding it. It's good to get an email for a required step in the process. If I upload a package, I get a few confirmation mails when the upload gets processed by various systems. This is good; if I do not receive such an email, that means something is wrong and I should investigate. I have not yet received such an LLM-based review mail, so don't know whether the following applies, but: If the LLM is invoked before the "your package is now in NEW" mail is sent out, then it could reasonably be added to that email as a "no guarantees but this is what our LLM flagged, feel free to ignore" paragraph in that email. This would not be unreasonable. If the LLM is invoked *after* that email, sending out an email for just the LLM output is noise and should be avoided. In such a case, I would think it makes more sense to have an "LLM analysis" section on the NEW queue website which people can ignore if they want to. -- "I never had a C in history!" "Yeah, but there was so much less of it when you were my age!" -- Joe Brockmeier recounting a conversation with his father, cfgmgmtcamp 2026, Ghent
