If your agent gets flooded – detect the flooding.
If code gets obfuscated – detect the obfuscation.
If ETW gets silenced – detect the silence.
If the EDR gets killed – detect the killing.
If logs get cleared – detect the clearing.
The act of hiding is often more suspicious than what’s being hidden.
It’s like a surveillance camera going black or freezing.
That is the signal.
I’ve been doing this successfully for years.
I detect obfuscated crap all the time.
People ask, “What is it?”
I say, “No fucking clue. Could be:
– a Themida-packed sample with a Microsoft copyright,
– a UPX-packed ELF with a 1-char filename,
– a PowerShell script that looks like static noise, or
– a fake svchost.exe with no Microsoft copyright.”
I don’t need to know what it is.
It’s obviously shady.
That’s enough to detect it – and deal with it.
There’s a Chinese saying that fits perfectly: 欲蓋彌彰
The more you try to hide it, the more obvious it becomes.
The largest collection of malware source code, samples, and papers on the internet.
Password: infected
That appeared to be untrue as vx-underground, ran by a team of volunteers, started in 2019 ([Archive] web.archive.org/web//vx-underground.org) right when a few crisis in my life came together at the same time.
So here are the links for my archive as they are great content for both Red Teams and Blue Teams on many things cyber security related:
Lot’s of references by [Wayback/Archive] Parsia to great posts by [Wayback/Archive] Raymond Chen mainly on security issues that are not: there is only a vulnerability when you get from the other side of the outside of the airtight hatchway to the inside, not when you are already inside.
Arthur: But can’t you think of something?!
Ford: I did.
Arthur: You did!
Ford: Unfortunately, it rather involved being on the other side of this airtight hatchway—
Arthur: oh.
Ford: —that’s just sealed behind us.