TrickBot: New Injects, New Host

What’s in the Name: Call it IcedID or TrickBot? Tell that to a security researcher (Arsh Arora in this case) and watch them RANT

(Gar-note: today’s blog post is a guest blog from malware analyst, Arsh Arora…) 

Today’s post starts with an interesting link from Dawid Golak’s Medium post: “IcedID aka# Bokbot Analysis with Ghidra” which mentions that IcedID is dropping TrickBot. Although the article is about IcedID, it gets confusing quickly, because the researcher focused on finding artifacts for IcedID instead finds TrickBot artifacts. A big question for the security industry still remain is to how to classify the malware from the originator or the binary that is being dropped. We followed up on the sample he mentioned and saw the same thing.  This is definitely Trickbot.

First Stage – Sample Collection from Virus Total Intelligence

In the “AnyRun Analysis” linked to by Dawid, the TrickBot binary was downloaded from “54.36.218[.]96 (slash) tin[.]exe

Fig 1: TrickBot Sample

Second Stage – Sample Execution

After the execution in a virtual environment, I was able to see TrickBot behavior similar to what we have documented in the past in our post “Trickbot’s New Magic Trick: Sending Spam“:

A large number of config files and dlls were loaded into the Roaming/netcache/Data, a  unique behavior of the TrickBot binary.

Fig 2: Configs and Dlls Loaded

Third Stage – Open Firefox and visit different Bank website

It is often the case that to get any banking trojan to co-operate with the researcher, some initiation from the researcher side is needed. Due to past experience, I have learned that one needs to open up a browser and visit different bank websites to activate the banking trojan. The trojan resists until instigated by visits to these pages. I visited close to 20 different bank websites and was able to obtain injects from 7 of those bank websites. The injects and admin login panels of the websites are as follows.

Name of  Bank
Admin Login Panel
Bank of
AS9009, Prague
AS9009, Prague
AS48282, RU
AS50673, NL
AS50673, NL
53 Bank
AS50673, NL
When infected, viewing the source code while visiting one of the banks is all that is needed to identify the data exfiltration destination.  Some examples follow from this infection run:


Fig 3: BoA Web Inject


Fig 4: Chase Web Inject

Fig 5: BoA and Chase Admin Panel


Fig 6: Citi Web Inject

Fig 7: Citi Login Panel


Fig 8: USAA Web Inject


Fig 9: WellsFargo Web Inject

Fig 10: WellsFargo Admin Panel


Fig 11: PNC Web Inject

Fig 12: PNC Admin Panel

53 Bank

Fig 13: 53 Bank Web Inject

Fig 14: 53 Bank Admin Panel

For more details please contact Arsh Arora (ararora at or Gary Warner (gar at at UAB. Please note:  Arsh is defending his PhD this summer and looking for new opportunities.

*** This is a Security Bloggers Network syndicated blog from CyberCrime & Doing Time authored by MalwareSecrets. Read the original post at:

Cloud Workload Resilience PulseMeter

Step 1 of 8

How do you define cloud resiliency for cloud workloads? (Select 3)(Required)