@RyanM That pattern looks like it's already caught by Blacklisted website in answer and Blacklisted website in body Append -force to the command if you really want to add this pattern.
@SmokeDetector That's an awful lot of flags to be cast in 12 seconds for a non-autoflagged post, especially by people who don't appear to be pingable here or in SOCVR. What's up with that?
is there a robustness problem with the logic? I'm not too familiar with these parts but I guess if it relies on being able to ask itself what it has done (rather than try to keep track within the flagging logic itself) it could overflag if the communication breaks down
this spammer seems to conspicuously limit their spams to three per domain metasmoke.erwaysoftware.com/domains/53834 but not enough samples to really establish a pattern
[ SmokeDetector | MS ] Bad keyword with a link in answer, link at end of answer, potentially bad ip for hostname in answer, potentially bad keyword in answer, potentially bad keyword in username (145): Getting Stripe to work with Commerce by Maria Costa on craftcms.SE
@JeffSchaller That pattern looks like it's already caught by Potentially bad keyword in body and Potentially bad keyword in answer Append -force to the command if you really want to add this pattern.
[ SmokeDetector | MS ] Link at end of answer, potentially bad asn for hostname in answer, potentially bad ns for domain in answer, potentially problematic ns configuration in answer, username similar to website in answer (130): Main Menu not showing in wordpress by FinancialGuide on stackoverflow.com
Autoflagged FP: flagged by @SmokeDetector, @Rob, @Stormblessed
[ SmokeDetector | MS ] Bad ip for hostname in body, bad keyword in body, blacklisted website in body, body starts with title and ends in url, pattern-matching website in body (487): For a few of us by Tim Zee on stackoverflow.com
I guess SD does not follow per-site-metas, only main sites? If I look at the spam posts from MathOverflow meta mentioned in this room, it seems that all of them were reported manually: chat.stackexchange.com/…
@MartinSleziak It's something tripleee brought up recently. To my knowledge, we used to fully scan all meta sites the same way we scanned main sites (with some adjustments to prevent too much api quota drain) - but now we're... Not sure, I guess.
I believe the intention was definitely to scan them like main sites.
One could also argue that metas are more attractive for spammers, it all depends on the way we look at the issue.
Less users visit meta, so there are less users who will flag the post as spam. It will probably survive a bit longer than on the main site, it is more likely to get indexed by Google and other search engines.
Posting on meta will alert the site's mods, so unless there are many meta activities (i.e. spammer on MSO using SOfT) or mods are reluctant to check them, spam on meta usually won't last that long.
@AndrewT. You probably mean the moderator inbox? IIRC only questions from meta go to the moderator inbox. In this case (and a few other cases I saw), spam can be posted as an answer.
In any case, one hour went without new spam post on MathOverflow Meta - so perhaps we won't see this spammer again.
After the spam answers are gone, probably the most annoying thing is that some ancient questions were bumped. (But perhaps this can be seen as a good thing - after they were bumped, maybe some people will check whether they needs some update, retag, etc.)
