Hot Take Missed script malware by signature analysis

Status
Not open for further replies.
MS Defender - Default Settings

1.png

2.png
 
MS Defender - Default Settings


We are missing the first part from goo.su that executes the disable defender functions). The one that user copy-pastes.

So almost all of AVs detect the downloaded exe; what really matters, which one will detect the copied command before execution filelessly.
Not many, such detection will be Nightmare on Windows Street and will be jam packed with false positives.
 
The real problem MD has not reach such stage to detect the payload; the script execution disabled MD first.
I do like to watch a video testing the resilience of major AVs to deactivation of fileless scripts.
OK, Shadowra sent me the video now.

The video advises users to disable the AV because “files are completely clean”.

Though Defender still could be disabled (or exclusion could be added), the attacker did not apply this method.

This social engineering can be applied with any product.
IMG_3136.jpeg
 
.SU domain, man it's been awhile but that has always been hella shady. So much bad stuff used to happen with that domain but I thought it died! Maybe not.
This is the first clue for the users. Usage of .su TLD are mostly being utilized by Cybercriminals.
Hard to spot if target users are kids/gamers which don't have an idea what it is.
 
Yeah, the attacker relies on videos with loads of tags. They need “stable” architecture with very little control and this URL shortener (goo.su) probably has inefficient abuse reporting and very little control. Otherwise the attacker will have to edit all video descriptions on daily basis once a URL shortener blocks their accounts.

However, the attacker also relies on GitHub.

The account after I (and potentially other people before me) reported the abuse has been deleted.

It is unlikely that goo.su links can be updated so that’s a whole bunch of work, gone down the drain. 🤷🏻‍♂️
 
The real problem MD has not reach such stage to detect the payload; the script execution disabled MD first.
I do like to watch a video testing the resilience of major AVs to deactivation of fileless scripts.
MD Home is a sitting duck against such disabler scripts. The Enterprise edition fares well. I believe this is a deliberate oversight by Microsoft.
 
It is not an oversight, disabling Defender requires admin privileges. It is a common philosophy that any action authorised by the admin is the responsibility of the admin.

So this is not considered a programatic attack, it falls into the admin with malicious intent category.

Microsoft doesn’t protect against that.

Automating Defender management through PS can be used for legit purposes, but I am not sure why Microsoft doesn’t ship a setting “Enable automation through PowerShell CMDLets” so users not in a need of it can switch it off.

Anyway, other cmdlets like the WMI-related ones are soon gonna be phased out so an entire class of persistence will soon disappear.
 
Another sample; B blocked connection of mshta.exe to a url; typed such url in address field and enter: downloaded an hta file (initially was blocked by SS, and after download MD asked to upload for analysis); however, scan by MD is negative.



Uploaded to VT (detected by K and Avast).

View attachment 292802

Nice job by GenD K and B to detect this RHADAMANTHYS' latest persistence mechanism truly showing their effectiveness against zero day malware.
 
Its flag was removed later, do not why! Anyway is was inaccurate one (phishing).
View attachment 292814
Here is another post about AVG missing a malware
 
Here is another post about AVG missing a malware
I am now more concerned about protective mechanisms against deactivation by malware than missing few samples; could not find a comparative video for major AVs in this regard.
 
Status
Not open for further replies.