We use “Require Approval” Watched Words to catch persistent spam domains.
Spammers are now posting anodyne responses with no triggering words in them and then editing their posts to contain the spam link to a domain that is on our Watched Words list.
Is it intended that Watched Words can be circumvented by post editing?
I’m afraid that is currently the expected behaviour, though a Contribute > Feature request was split off from the topic you linked above to ask for it: