They're trying really hard to make their AI-plagiarized slop "news" site happen. You'll see posts linking to primestories24 [dot] com but they're AI slop regurgitating likely legit news articles. Every "reporter" has some generic profile that is most likely the system prompt used for everything they "write".

If you see a post to that domain, report it as spam. The accounts all look similar to this:

This list isn't comprehensive.

Update: They are now using sockpuppet alts to vote/boost their posts. Just found @Babrah28@ttrpg.network that way and, surprise surprise, it was registered the same time as the latest batch.

Note: Technically the above is violating rule 7 if you consider spambots to be members. I don't, but if this is a problem, let me know, and I'll remove that part.

The domain was registered in May of 2026 and is roughly when these accounts started popping up.

   Domain Name: PRIMESTORIES24.COM
   Registry Domain ID: 3105727536_DOMAIN_COM-VRSN
   Registrar WHOIS Server: whois.web4africa.net
   Registrar URL: http://hostafrica.com/
   Updated Date: 2026-06-19T10:01:29Z
   Creation Date: 2026-05-31T12:16:00Z
   Registry Expiry Date: 2027-05-31T12:16:00Z
   Registrar: Host Africa (Pty.) Ltd.
   Registrar IANA ID: 664
   Registrar Abuse Contact Email: abuse@web4africa.net
   Registrar Abuse Contact Phone: +1.6465850088
   Domain Status: clientTransferProhibited https://icann.org/epp#clientTransferProhibited
   Name Server: DAN1.HOST-WW.NET
   Name Server: DAN2.HOST-WW.NET
   DNSSEC: unsigned

Admins, I'm aware Lemmy is severely lacking in moderation capability, but one thing it does have is domain blocking. That prevents posting to those domains and it now seems to prevent links to those domains from federating in.

you are viewing a single comment's thread
view the rest of the comments
[–] [S] 1 point 4 months ago* (last edited 4 months ago) (3 children)

It does work on URLs. The https part gets stipped out because of how the regex

Oh, nice. For some reason I assumed it was only domains.

It always worked on federated content?

I know it has for a good while but I forget when it was applied to inbound federation. I think it was that Mastodon/Misskey spam wave I was thinking of but don't recall which version was running at the time and how I wished it would stop content to those domains from federating in. Nice contribution, BTW.

  • source
  • parent
  • hideshow 3 child comments
  • [–] 3 points 4 months ago (2 children)

    Thanks! I have also floated the idea of a regex blocklist, but Nutomic shot that down because 'it can be done with a plugin'. Can't wait until 1.0 so Lemmy can have feature parity with a 2006 forum.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 2 points 4 months ago* (1 child)

    I think that a heavyweight automod feature that works something like SpamAssassin, where one builds a training corpus of known spam and known ham, and then one writes a bunch of rules (e.g. "domain is less than X days old", "account has less than N votes", etc) and then runs it on the training corpus to auto-generate score weights for the rules, and then runs the rules on each post to provide an estimated "spam probability score" on given posts would probably scale better.

    Doesn't need to be built into Lemmy or Mbin or PieFed, and running as an automod bot, it'd be able to run on all of those platforms, share effort among them.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 3 points 4 months ago

    I don't disagree, and the plugin system is cool as it'll enable people to build things like that. I think SJW even has something like that already. But I can't begin to tell you how much work a simple text filter would have saved me as an admin.

  • source
  • parent