This is a proof of concept meant to partially automate labels, I'd like to request for comment (given that I also took the liberty of adding a bunch of new labels) first before testing / working further on this.

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 77.10%. Comparing base (c113768) to head (e40754a).

@@           Coverage Diff           @@
##             main    #1044   +/-   ##
=======================================
  Coverage   77.10%   77.10%           
=======================================
  Files          28       28           
  Lines        3324     3324           
=======================================
  Hits         2563     2563           
  Misses        761      761           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

I'm not quite sure if the project is that big, that we have to automatically label our PRs. For me, that feels like this gets relevant once different persons maintain different subsets of the project, which isn't the case for uhyve. But if you feel the need, I'll not oppose it as long as it is just labelling.

that we have to automatically label our PRs

I see your point, but my argument for that is that I tend to look up past PRs or issues (and there are often duplicate issues) in retrospect. I'd say that this is more about bookkeeping and information retention – in the context of 900+ PRs, the project may as well be considered big enough. On the other hand, labeling is manual effort and this just automates like 30%-40% of it. Ballpark numbers.

For big projects with components maintained by different people, I'd use CODEOWNERS which is not the case here.

Anyway, I'll test it on my fork and mark it as Ready for review whenever I find the time.

As some time has passed, do you want it to be merged?

As some time has passed, do you want it to be merged?

Let's just go for it? :D