Hi. Thanks to the moderators for reopening this topic on my request 
It’s been 5 years, and I feel the need to restart the conversation, saying, please, this is more important now than ever. I feel that this topic’s contents are still valuable, so let’s continue the conversation.
Please have a look at my post above (the one I’m replying to here) to see the use cases I detailed there.
The “suspected spammer” scenario is one that I am encountering daily because spammers are getting more intelligent to react to Discourse’s evolving anti-spamming features. So now they typically engage in a human fashion, not just bots, and try to live in the community normally for a while before starting to drop their SEO links, usually through post edits, not new posts.
It would be so relaxing to be able to just signal to Discourse: this guy looks fishy, tell me if he does anything such as edit his own posts adding links.
Even if the idea of flagging users to be followed temporarily is too much, I would propose another straight-forward feature:
if any user edits his own post to add a link, or change an existing link there, this edit goes into a moderator queue to be examined.
This could be a preliminary screening (edit isn’t effective until approved by a mod) or just a warning (edit is made, but moderators can then go check it, and somehow mark it as already checked so that not all the other moderators waste time doing it).
But I would still prefer that admins can generically follow users, due to the other use cases I listed above.