It's a proposal to add a `code` property to Error objects similar to the recently added `cause`... works essentially the same way. It is currently stage 1 but will be up for stage 2.7 advancement at the plenary next week. There are currently no indications of any blockers.
I have an implementation (with corresponding in-progress flag gate) for v8 locally ready to go but obviously a bit too early to land. Wanted to see if anyone was opposed to at least getting the merge request opened even if it sits for bit.