As this is currently implemented, users don't have a way to disable this filtering behavior other than manually modifying this code. livecheck blocks for autobumped formulae/casks regularly break, so we need a way to be able to run these checks locally for debugging/fixing. A CLI option to also check autobumped packages (e.g., --autobump) would be a simple addition.

Beyond that, there may be something to be said for also offering an environment variable that would act like --autobump is always used (e.g., HOMEBREW_LIVECHECK_AUTOBUMP). My situation may be a bit rare (since I do a lot of livecheck block maintenance) but I almost never want to skip autobumped packages, so I would greatly prefer to not have to add a CLI option a lot of the time (and having to completely re-do long runs if I forget).

I tested this in a variety of scenarios (with/without --autobump, with/without HOMEBREW_LIVECHECK_AUTOBUMP=1, normal/debug/JSON runs, only autobumped or not autobumped packages, a mix of autobumped and not autobumped packages, formulae and casks, first-party and third-party taps, etc.) and it appears to work as expected. Nice work and thanks for adding --autobump and HOMEBREW_LIVECHECK_AUTOBUMP (that'll save me some effort on a regular basis).

In case there's any question, this change shouldn't affect brew bump because it has its own logic (using some of the Livecheck methods). A quick test with HOMEBREW_TEST_BOT_AUTOBUMP=1 brew bump a2ps works as expected. Similarly, it shouldn't affect the core/cask autobump workflows because those use brew bump.

I added a few suggestions but the only blocker is the ENVS order, as that's causing a test to fail.

Thanks, @issyl0!