I'm seeing some unmasked badness in this PDR1 COSMOS coadd on lsst-dev:

repo: /datasets/hsc/repo/rerun/DM-13666/UDEEP

dataId: tract=9813, patch='4,4', filter='HSC-R'

It looks like ringing of some kind (a bit similar to what I saw on DM-12985), but I'm not sure if it's due to the nearby bleed trail, the nearby REJECTED region, or something else. (As a side note, I don't see any other mask plane explaining why it was REJECTED, and that might be a problem of its own).

This region is masked BRIGHT_STAR, but we shouldn't be relying on that to catch things like this.

First step to investigate this is probably to look at the input epochs and see if it appears there.

A screenshot from Firefly (which is super nice for detailed inspections of the mask plane overlays) is attached.