This is a lot of good news for discoverability of current policy.

Re your last paragraph, I clearly need to explain what that "example account" of mine has been doing, because my Q + A posts here were aiming at certain aspects that I wanted to discuss only, while suppressing other aspects sufficiently covered elsewhere. I will describe three or four different types of spam accounts in this comment thread.

Spam account type A keeps posting questions or answers whose text is blatantly promotional and off-topic for the site; if the posts contains hyperlinks, they are equally promotional and off topic for the site (but naturally placed within the promotional post itself).

Spam account type B keeps posting very low quality questions or answers that are nominally on topic, perhaps AI generated, or lazily googled up or consist of generic, unhelpful frame challenges. The true purpose and payload of the disingenious posts are the HYPERLINKS that are blatantly promotional and randomly placed throughout the posts, in no relation to the post text.

Spam account type C keeps posting very low quality questions or answers that are nominally on topic, perhaps AI generated. All their promotional content within our site (almost certainly complete with hyperlinks to the outside) is located in their account profile to which each their post, implicitly and automatically links.

Spam account type D keeps posting very low quality questions or answers that are nominally on topic, perhaps AI generated. All their promotional content within our site is their account name which is chosen to be the (relatively unknown but globally present) brand name they want to promote.

Thanks for these examples! Spam account A is the easiest of this lot: if it's only a spammer (no valid posts) you can delete that sucker, or if you think it's redeemable or it has valid posts too, you can suspend. Spam account B sounds like at least a suspension (and editing or deleting the low-quality posts). C and D are where things are not as clear as we'd like, and I thank you for raising this.

All spam account types named so far have many common properties: no trace of any honest contribution to the site, lots of posts, receiving a steady stream of flags and downvotes without any genuine upvotes, an obvious commercial agenda, and a distinct "robotic" feel where they don't seem to truly read anybody else's posts, nor their own.

However, types C and D use indirection and their posted texts, while being pure useless clutter, does not immediately radiate their business agenda. If type C wins any exposure at all, I suppose they win it through SEO rather than through the human users of Codidact sites.

The example account I asked this OP about started out as a combination of the D + B types. They received a formal warning specifically about those randomly placed hyperlinks; after receiving the warning, they immediately stopped to include the hyperlinks in their subsequent posts, becoming a pure type D. Unsurprisingly, their account name exactly matched the domain name of every single hyperlink they posted while they were still D + B; this proves that the account name couldn't have been coincidental. A pure D with no such history would be harder to prove in the wild but still notionally possible.

Oh, that sounds like a real nuisance! I just took a look at recent flags and it looks like those accounts have been nuked (thanks). I'll see if someone with DB access can spot any common factors (like IP). Next time you get one, feel free to escalate the flag -- admins have a "stronger" nuke option, but if they're using VPNs or are just scattered, that won't help. Doesn't hurt to look, though!

Interestingly, type C also has a variety that doesn't even have to post within our network. It has been described here. Somebody might say, this variety is no spammer, they don't post to pages anyone cares about. But it is still parasitic behavior ultimately aimed at gaining exposure (somewhere).

Thanks for all your support - I guess I'm currently fine and I understand that I could escalate a flag if the scale required it.

I agree that's parasitic behavior and we should make a clear statement about it. Thanks for linking to that staff answer -- good to have on record. And please do ping us or escalate flags if these accounts are frustrating you! We appreciate our volunteer moderators and never want this to be a chore.

Don't worry, I am not frustrated in the least, nor at risk of ever getting. I'm helping out with moderation of a single microscopic site with nearly zero traffic ever, both legit and spam. It's an interesting life experience nevertheless. I find Codidact's software surprisingly robust and feature rich and thoughtfully designed to help me in every situation I might encounter at the current scale. It's just that I don't use Discord a lot and so I tend to ask the few policy questions I might ever have at meta, to learn from more experienced moderators, users, or staff, and maybe more importantly, to improve the discoverability of the policy for everyone.

I agree with asking policy questions on Meta where they are more visible and discoverable! I've tried to find needles in the Discord haystack, and it's very hit-or-miss at least for me. And nobody's going to casually notice it after it scrolls out of view there.