opencode open source: the bottom line with community PRs are they are 99% of the time slop. even in the case that it looks like a good PR, there's suspicious things. this takes the team's time. we have to exactly repro the bug (because there's no steps in the issue/PR) there's a UI change with no before/after screenshots/video there's terrible code - we'd need to pull the code, refactor/cleanup, verify after there's tests that don't test anything - e.g. they always go green, even after you intentionally add the bug back at the end of the day, most the PRs are NOT worth the time it takes to properly go through it. in the case of an abysmal implementation we will instead just do the fix/feature ourselves from scratch. that being said - good OSS people still stick out, and their contributors get merged, and they get more trust, to be able to do bigger contributions. even when we merge community contributions they don't own the code once it merges - what if there's a fatal bug and we don't totally understand the new code? its completely fine to go disspear for a month as an OSS contributor. but that is risk we have to account for. at the end of the day the team has to tend to the garden and enforce some level of quality, sometimes this includes not merging PRs that look 80% good to go a lot of PRs will 'fix thing URGENT IT BREAAKS FOR EVERYONE ROLL THING BACK NOW', but then the linked 'fix' community PR breaks 1000s of other things.