AR20-239C: MAR-10257062-1.v2 – North Korean Remote Access Tool: FASTCASH for Windows

Original release date: August 26, 2020



This report is provided “as is” for informational purposes only. The Department of Homeland Security (DHS) does not provide any warranties of any kind regarding any information contained herein. The DHS does not endorse any commercial product or service referenced in this bulletin or otherwise.

This document is marked TLP:WHITE–Disclosure is not limited. Sources may use TLP:WHITE when information carries minimal or no foreseeable risk of misuse, in accordance with applicable rules and procedures for public release. Subject to standard copyright rules, TLP:WHITE information may be distributed without restriction. For more information on the Traffic Light Protocol (TLP), see



This Malware Analysis Report (MAR) is the result of analytic efforts between the Department of Homeland Security (DHS), the Federal Bureau of Investigation (FBI), and the Department of Defense (DoD). Working with U.S. Government partners, DHS, FBI, and DoD identified Remote Access Tool (RAT) malware variants used by the North Korean government. This malware variant has been identified as FASTCASH for Windows. The U.S. Government refers to malicious cyber activity by the North Korean government as HIDDEN COBRA. For more information on HIDDEN COBRA activity, visit https[:]//www[.]

FBI has high confidence that HIDDEN COBRA actors are using malware variants in conjunction with proxy servers to maintain a presence on victim networks and to further network exploitation. DHS, FBI, and DoD are distributing this MAR to enable network defense and reduce exposure to North Korean government malicious cyber activity.

This MAR includes malware descriptions related to HIDDEN COBRA, suggested response actions and recommended mitigation techniques. Users or administrators should flag activity associated with the malware and report the activity to the Cybersecurity and Infrastructure Security Agency (CISA) or the FBI Cyber Watch (CyWatch), and give the activity the highest priority for enhanced mitigation.

This submission included two unique files. The first file is a malicious application, which can be utilized to inject a dynamic link library (DLL) into a remote Windows process. The second file is a malicious Windows DLL. The DLL contains two functions that can hook callbacks to the Windows application programming interfaces (APIs) “Send” and “Recv” within a targeted process. These hook functions are utilized to intercept traffic received by the target process. In received Financial Messages, the malicious functions will look for targeted Primary Account Numbers (PAN) to deliver a custom response. It appears the malware will target a system on a bank infrastructure, which is designed to process automated teller machine (ATM) transactions.

This updated report included an additional sample that is used by advanced persistent threat (APT) cyber actors in the targeting of banking payment systems. The sample is a man-in-the-middle bank transaction modification malware. Once the malware is injected into an executable, it takes control of the send and receive functions in order to identify, log, and modify ISO 8583 messages. ISO 8583 is an international standard for financial transaction card originated interchanged messaging. This functionality enables the actor to withdraw more money than is actually available. The malware specifically targets ISO 8583 Point of Sale (POS) system messages, ATM transaction requests, and ATM balance inquiries. The sample uses code from open source repositories on the Internet and modifies the parsing code to support Extended Binary Coded Decimal Interchange Code (EBCDIC) encoding. EBCDIC is a character encoding format like the more commonly ASCII.

For a downloadable copy of IOCs, see [STIX file].

Submitted Files (3)

129b8825eaf61dcc2321aad7b84632233fa4bbc7e24bdf123b507157353930f0 (switch.dll)

39cbad3b2aac6298537a85f0463453d54ab2660c913f4f35ba98fffeb0b15655 (switch.exe)

5cb7a352535b447609849e20aec18c84d8b58e377d9c6365eafb45cdb7ef949b (A2B1A45A242CEE03FAB0BEDB2E4605…)





Name switch.dll
Size 118784 bytes
Type PE32 executable (DLL) (console) Intel 80386, for MS Windows
MD5 c4141ee8e9594511f528862519480d36
SHA1 2b22d9c673d031dfd07986906184e1d31908cea1
SHA256 129b8825eaf61dcc2321aad7b84632233fa4bbc7e24bdf123b507157353930f0
SHA512 dfc1ad2cb2df2b79ac0f2254b605a2012b94529ac220350a4075e60b06717918175cff5c22e52765237b78ec4edffd6df20f333e28a405a4339a10288158e7fc
ssdeep 3072:lUGDXTpE8AKDKDOf+8ZagCfG4aAzFdIARrhxg6/ZpDA:+GDXTpFDKDMZagX4aAB2Cg6hpD
Entropy 6.454745
Antiy Trojan/Win32.Tiggre
Avira TR/Spy.Banker.pubvd
BitDefender Trojan.GenericKD.32541173
ClamAV Win.Trojan.Alreay-7189205-0
Comodo Malware
ESET a variant of Win32/NukeSped.GA trojan
Emsisoft Trojan.GenericKD.32541173 (B)
Ikarus Trojan.Spy.Banker
K7 Riskware ( 0040eff71 )
Lavasoft Trojan.GenericKD.32541173
McAfee Trojan-Banking
NANOAV Trojan.Win32.NukeSped.gexoae
Sophos Troj/Banker-GYS
Symantec Trojan Horse
TrendMicro Backdoo.62DC2502
TrendMicro House Call Backdoo.62DC2502
VirusBlokAda BScope.TrojanBanker.Agent
Zillya! Trojan.NukeSped.Win32.183
YARA Rules
  • rule CISA_10257062_01 : ATM_Malware
           Author = “CISA Code & Media Analysis”
           Incident = “10257062”
           Date = “2019-09-26”
           Last_Modified = “20200117_1732”
           Actor = “n/a”
           Category = “Financial”
           Family = “ATM_Malware”
           Description = “n/a”
           MD5_1 = “c4141ee8e9594511f528862519480d36”
           SHA256_1 = “129b8825eaf61dcc2321aad7b84632233fa4bbc7e24bdf123b507157353930f0”
           $x3 = “RECV SOCK= 0x%p, BUF= 0x%p, LEN= 0x%08X, RET= %08X, IP= %s, Port= %d” fullword ascii
           $x4 = “init_hashmap succ” fullword ascii
           $x5 = “89*(w8y92r3y9*yI2H28Y9(*y3@*” fullword ascii
           ($x3) and ($x4) and ($x5)
ssdeep Matches

No matches found.

PE Metadata
Compile Date 2019-06-22 01:59:31-04:00
Import Hash 0ab159bd939411cb8df935bd9e7b5835
PE Sections
MD5 Name Raw Size Entropy
00f8301c11847b70346d6271098d8f1c header 1024 2.296500
c3bee35076d728ce32b67f5bc66587f3 .text 84992 6.641787
6b094443cad879acc7285f991243ddb0 .rdata 17920 5.170073
11060bd3e49075b78be8670ff46d9a48 .data 7168 4.275765
3637e0cd32608b060e308fdd9742ea97 .reloc 7680 4.792696
Microsoft Visual C++ DLL *sign by CodeRipper

This file is a malicious Windows 32-bit DLL. Upon execution, it attempts to read the file “c:\tempinfo.dat”. Analysis of this implant indicates the encrypted file “info.dat” will contain targeted PAN numbers, which are expected to be contained within transactions possibly originating from ATM systems. Analysis indicates the malware decrypts “info.dat” utilizing what appears to be the AES encryption algorithm. The key utilized for this decryption is displayed below:

–Begin Decryption Key–


–End Decryption Key–

The decrypted contents of “info.dat” are then parsed. Sub-components of the file are then further decoded using a hard-coded rotating XOR cipher (Figure 1). The data used as the rotating XOR cipher key is displayed below:

–Begin Rotating XOR Cipher Key–


–End Rotating XOR Cipher Key–

This application will not run without the file “info.dat”, which was not available at the time of analysis.

Upon execution, the malware creates the directory “C:tmp_DMP”. The malware will use this location as a working directory on the targeted system. The malware will store run time logs within this folder. When executed, the malware will create a log file with the following file name format “c:\tmp\_DMP\TMPL_%d_%d.tmp” in this folder and stamps it with the data “HK-Start”.

This binary contains two functions, which provides context to the malware’s purpose and capability. Analysis indicates this DLL is injected into a targeted process. In order to capture and analyze incoming network traffic, the malware hooks the “Send” and “Recv” Windows API within a targeted process. One of these functions, located at offset “0x00004f60”, appears to search for incoming network traffic for “x200” Financial Request Messages, such as the type that may be generated from an ATM banking system. When the malware captures data it uses the “getpeername” API to get the IP address of the connected host. It then converts this IP address to integer value using the “ntohs API”. If the integer value of the IP address matches either “16843029” or “33620245” the malware will search it for a “Financial Request Message” (Figure 6). If not, it will process the incoming data as normal, however it still attempts to log it to a file named “c:\tmp\_DMP\TMPL_%d_%d.tmp” in the format RECV SOCK= 0x%p, BUF= 0x%p, LEN= 0x%08X, RET= %08X, IP= %s, Port=.

Upon receipt of one of these Financial Request Messages, this structure will create a log file that is named with the following format: “c:\tmp\_DMP\TMPL_%d_%d.tmp”. The format of the data logged in this log file will be as follows:

–Begin Logged Message Data–

Message(msg=%d, ct=%d, pc=%d, sd=%d, pan=%s, date=%s)

–End Logged Message Data–

Upon receipt of a Financial Request Message the malware will decode a portion of the data, which was AES decrypted from the file “info.dat” to see if portions of it match the incoming Financial Request Message (Figure 3). Although the file “info.dat” was not available for analysis, it appears the malware is ensuring the PAN numbers of the incoming message match one of the PAN numbers contained within “info.dat”.

Static analysis indicates the malware utilizes an encrypted file named “blk.dat”. This file is expected to contain a blacklist of ATM transactions, which will be denied by the hook function (Figure 2). This file was not available for analysis.

When the malware receives a request from an ATM, if it contains a PAN number configured in info.dat (Figure 3) and it is not on the blacklist in “blk.dat”, the malware will craft a response and send it to the ATM system (Figure 4). It appears the response to the ATM will allow the transaction to proceed and potentially allow the hackers to illegally withdraw money. If the transaction is hijacked and approved, the malware records this success in the encrypted log file “suc.dat”.

If the transaction is rejected, because it is on the blacklist in “blk.dat”, this error is logged to the file “err.dat”. If the transaction does not contain a configured PAN or a transaction on the blacklist, the malware will pass it on as normal to the targeted application. When the malware receives an identified Financial Request Message, it will log it to a file with the name format “c:\tmp\_DMP\TMPL_%d_%d.tmp”. The message itself will be logged into this file with the format “Message(msg=%d, ct=%d, pc=%d, sd=%d, pan=%s, date=%s)”.

The actual response back to the ATM system will be logged into a file with the filename format “c:\tmp\_DMP\TMPL_%d_%d.tmp”. The format of the data written to this file will be send socket=0x%X, ret=%d, err=%d.

Analysis indicates the Send API is hooked with a function that uses the “getpeername” IP address of the connected host. The IP address of the host is converted using “ntohs” and if it matches one of the values “16843029” or “33620245” the sent traffic will be logged in a file named “c:\tmp\_DMP\TMPL_%d_%d.tmp”. The format of the sent data logged is SEND SOCK= 0x%p, BUF= 0x%p, LEN= 0x%08X, RET= %08X, IP= %s, Port= (Figure 7). Static analysis indicates successful hooks made to the “Send” and “Recv” APIs within the target process will be logged in a file named “c:\tmp\_DMP\TMPL_%d_%d.tmp” with the format “g_hook_flag = %d”.


Figure 1 - Cipher used when decoding data in "info.dat".

Figure 1 – Cipher used when decoding data in “info.dat”.

Figure 2 - API "Recv" hook checking for incoming Financial Request Message for a targeted PAN.

Figure 2 – API “Recv” hook checking for incoming Financial Request Message for a targeted PAN.

Figure 3 - The malware searching for targeted PANs.

Figure 3 – The malware searching for targeted PANs.

Figure 4 - Malware crafting and sending responses to the ATM.

Figure 4 – Malware crafting and sending responses to the ATM.

Figure 5 - Hook function either searching network traffic for Financial Message or logging it and sending to the "RECV" API.

Figure 5 – Hook function either searching network traffic for Financial Message or logging it and sending to the “RECV” API.

Figure 6 - "RECV" Hook API function checking if the connected host is one of the two IP addresses.

Figure 6 – “RECV” Hook API function checking if the connected host is one of the two IP addresses.

Figure 7 - Logging outbound traffic to the two specific IP addresses.

Figure 7 – Logging outbound traffic to the two specific IP addresses.




Name switch.exe
Size 67448 bytes
Type PE32 executable (GUI) Intel 80386, for MS Windows
MD5 89081f2e14e9266de8c042629b764926
SHA1 730c1b9e950932736fc4b02cbdb4e4e891485ac2
SHA256 39cbad3b2aac6298537a85f0463453d54ab2660c913f4f35ba98fffeb0b15655
SHA512 bbb5aa4d8e7a011daff71774ee9c74fa4d14627de1c25e0437c879bd1cd137223d5c2fb20fd101a511a95e59d91ea884b0947229ee67e40a4a24350573fb9e54
ssdeep 768:aQ1PWoWzXyjJsTKJUniYs1pdLn4nDT622YuYDIhscWTJqLPNofEDy9nAXmIEHbKa:aQ5WDziX+nD0LWT6FYZDgs5ULPIJEYp
Entropy 6.396614
Ahnlab HackTool/Win32.Injector
Antiy Trojan[Banker]/Win32.Alreay
ClamAV Win.Trojan.Alreay-7189192-0
Comodo Malware
ESET a variant of Generik.CWSORYC trojan
Emsisoft Gen:Variant.Ursu.634943 (B)
Ikarus Trojan.Inject
K7 Riskware ( 0040eff71 )
McAfee Trojan-Banking
Microsoft Security Essentials Trojan:Win32/LazInjector.DD!MSR
NANOAV Trojan.Win32.Alreay.geqrko
Sophos Troj/Banker-GYS
Symantec Trojan Horse
TrendMicro TROJ_NO.4FADD924
TrendMicro House Call TROJ_NO.4FADD924
VirusBlokAda TrojanBanker.Alreay
Zillya! Trojan.Alreay.Win32.96
YARA Rules

No matches found.

ssdeep Matches

No matches found.

PE Metadata
Compile Date 2018-06-13 02:17:06-04:00
Import Hash c9febdea3218b92a46f739082f26471e
PE Sections
MD5 Name Raw Size Entropy
cde81f1500263860f325ee8f80c483ce header 1024 2.497464
a8c0a36524287fef367821e833a68350 .text 38912 6.518662
e1c66ff8e5f0e1909e2691360c974420 .rdata 10752 4.878020
22783e6c2539d6828f3d42b030ca08e9 .data 4096 2.117927
81195ca9b22c050f79e44175e9e7150e .rsrc 512 5.105006
36571bcb45b1ae18dfcf7edc8c5c3d4a .reloc 3584 4.791228
Microsoft Visual C++ ?.?

This file is a malicious 32-bit Windows executable. It is a command-line utility. Static analysis indicates its primary purpose is to allow a user to inject a DLL into a remote process.




Name A2B1A45A242CEE03FAB0BEDB2E460587
Size 130560 bytes
Type PE32 executable (DLL) (console) Intel 80386, for MS Windows
MD5 a2b1a45a242cee03fab0bedb2e460587
SHA1 e9c9ef312370d995d303e8fc60de4e4765436f58
SHA256 5cb7a352535b447609849e20aec18c84d8b58e377d9c6365eafb45cdb7ef949b
SHA512 4ced785089832287d634c77c2b5fb16efb2147b75da9014320c98d1bc0933504bfba77273576c35b97548d25acb88a0f2944cbef6a78509f945a8502f8910da8
ssdeep 3072:j5KO2SQhF+VJbGHMjjNNyCkeZjDYJklGCx:oO2SQT+nGHADyAZjJwC
Entropy 6.431962
VirusBlokAda BScope.TrojanBanker.Agent
YARA Rules
  • rule CISA_3P_10257062 : HiddenCobra FASTCASH trojan
           Author = “CISA Trusted Third Party”
           Incident = “10257062”
           Date = “2020-08-11”
           Actor = “Hidden Cobra”
           Category = “Trojan”
           Family = “FASTCASH”
           Description = “Detects HiddenCobra FASTCASH samples”
           MD5_1 = “a2b1a45a242cee03fab0bedb2e460587”
           SHA256_1 = “5cb7a352535b447609849e20aec18c84d8b58e377d9c6365eafb45cdb7ef949b”
           $sn_config_key1 = “Slsklqc^mNgq`lyznqr[q^123”
           $sn_config_key2 = “zRuaDglxjec^tDttSlsklqc^m”
           $sn_logfile1 = “C:\intel\_DMP_V\spvmdl.dat”
           $sn_logfile2 = “C:\intel\_DMP_V\spvmlog_%X.dat”
           $sn_logfile3 = “C:\intel\_DMP_V\TMPL_%X.dat”
           $sn_logfile4 = “C:\intel\mvblk.dat”
           $sn_logfile5 = “C:\intel\_DMP_V\spvmsuc.dat”
           all of ($sn*)
ssdeep Matches

No matches found.

PE Metadata
Compile Date 2018-07-03 08:11:16-04:00
Import Hash 76e8a4f811b021cf503340a0077515cc
PE Sections
MD5 Name Raw Size Entropy
cbe7e7fdab96c22785fa8d7c03ca6b2b header 1024 2.429436
03d36f4d9ae3e002027c981c399ab8c6 .text 89600 6.630313
d1f983704c508544b315d577fe3563e1 .rdata 23040 5.215776
a4b79dca294053725e2b2091453d9d85 .data 8192 4.358771
d762ef71411860ae50212e14c0a5ba72 .rsrc 512 5.115767
2e4eb6056385f6f721d970cafe65bebe .reloc 8192 4.774185
Microsoft Visual C++ DLL *sign by CodeRipper

The file uses a configuration file, a black-list, and a series of log files:

–Begin files–
C:intelmyconf.ini: Configuration file that contains account numbers (encrypted) C:intelmyblk.dat: Black-listed account numbers (encrypted) C:intel_DMP_Vspvmlog_<PID>.dat: Logs general messages and errors.
Entry Format: [<YYYY-MM-DD HH:MM:SS.sss>][PID:<PID>][TID:<TID>] <Message>”] C:intel_DMP_Vspvmdl.dat: Logs API hooking/unhooking success and failure.
Entry Format:
Hook Success Entry: ‘Windows’
Hook Error Entry: ‘Linux’
UnHook Success Entry: ‘Acer’
UnHook Error Entry: ‘Lenovo’
C:intel_DMP_VTMPL<PID>.dat: Logs Send/Receive Message metadata
Entry Format:
Recv Entry: ‘recv – SOCK=<socket_id>, Addr=<IP>, Port=<Port>, pBuf=<data>, size=<datasize>’ Send Entry: ‘send – SOCK=<socket_id>, Addr=<IP>, Port=<Port>, size=<datasize>’ C:intel_DMP_VTMPR<PID>.tmp: Logs Received Messages
C:intel_DMP_VTMPS<PID>.tmp: Logs Sent Messages
C:intel_DMP_VTMPHSMS<PID>.tmp: Logs LocalHost ARQC sent messages C:intel_DMP_VTMPHSMR<PID>.tmp: Logs LocalHost ARQC received messages
C:intel_DMP_Vspvmscap.dat: Logs modified sent messages
C:intel_DMP_Vspvmsuc.dat: Logs modified sent messages metadata (encrypted)
–End files–

Upon attaching to a process, the sample will decrypt the encrypted config from the configuration file and read it into memory. Next, it will hook the processes send and recv winAPIs. When the “send” function is called, it will check to see if the port is 7029, if so, it will log the data and metadata in the above log files, if not it will just pass through calling send as the program normally would. When the “receive” function is called, it will check to see if the port is 7029, if so, it will wait for packets received from port 7029 and parse the following ISO8583 fields out of the incoming datagram:

–Begin fields–
–End fields–

Next, it checks the loaded configuration for the PAN. If it exists, it will continue processing, otherwise it will pass. Then it will check the blacklist file for the PAN. If blacklist contains ‘all’ or the PAN, will set the RESPONSE_CODE to 51 (Insufficient funds) in the response message. It looks for the following message types:

–Begin message types–
POS system message
ATM transaction request
ATM balance inquiry
–End message types–

Next it, constructs what appears to be an Authorization Request Cryptogram (ARQC) message:

–Begin format–
Contains the hardcoded string: “U8BFE0AE12F9000C1480B297BE43CAC97”
Sends to localhost on port 9990
Parses the response Authorization Response Cryptogram (ARPC) message
–End format–

Finally, it constructs and sends a ISO8583 response message.

When detaching from the process, the sample unhooks the “send” and “recv” WINAPI functions, returning them to their normal state. It will then overwrite the first 0x400 bytes of the in-memory DLL from the process, effectively cleaning up any trace of the sample.

The sample frequently uses code that is taken from GitHub with a few modifications in some cases. The sample uses code that is taken from to load the configuration file into memory in a hashmap, API hooking using Microsoft’s Detour library at and the ISO8583 parsing code is taken from (slightly modified to facilitate parsing of IBM037 formatted data).

The encryption that is used for all log/config files is likely an AES variant with the following keys:

–Begin keys–
–End keys–


CISA recommends that users and administrators consider using the following best practices to strengthen the security posture of their organization’s systems. Any configuration changes should be reviewed by system owners and administrators prior to implementation to avoid unwanted impacts.

  • Maintain up-to-date antivirus signatures and engines.
  • Keep operating system patches up-to-date.
  • Disable File and Printer sharing services. If these services are required, use strong passwords or Active Directory authentication.
  • Restrict users’ ability (permissions) to install and run unwanted software applications. Do not add users to the local administrators group unless required.
  • Enforce a strong password policy and implement regular password changes.
  • Exercise caution when opening e-mail attachments even if the attachment is expected and the sender appears to be known.
  • Enable a personal firewall on agency workstations, configured to deny unsolicited connection requests.
  • Disable unnecessary services on agency workstations and servers.
  • Scan for and remove suspicious e-mail attachments; ensure the scanned attachment is its “true file type” (i.e., the extension matches the file header).
  • Monitor users’ web browsing habits; restrict access to sites with unfavorable content.
  • Exercise caution when using removable media (e.g., USB thumb drives, external drives, CDs, etc.).
  • Scan all software downloaded from the Internet prior to executing.
  • Maintain situational awareness of the latest threats and implement appropriate Access Control Lists (ACLs).

Additional information on malware incident prevention and handling can be found in National Institute of Standards and Technology (NIST) Special Publication 800-83, “Guide to Malware Incident Prevention & Handling for Desktops and Laptops”.

Contact Information

CISA continuously strives to improve its products and services. You can help by answering a very short series of questions about this product at the following URL:

Document FAQ

What is a MIFR? A Malware Initial Findings Report (MIFR) is intended to provide organizations with malware analysis in a timely manner. In most instances this report will provide initial indicators for computer and network defense. To request additional analysis, please contact CISA and provide information regarding the level of desired analysis.

What is a MAR? A Malware Analysis Report (MAR) is intended to provide organizations with more detailed malware analysis acquired via manual reverse engineering. To request additional analysis, please contact CISA and provide information regarding the level of desired analysis.

Can I edit this document? This document is not to be edited in any way by recipients. All comments or questions related to this document should be directed to the CISA at 1-888-282-0870 or CISA Service Desk.

Can I submit malware to CISA? Malware samples can be submitted via three methods:

CISA encourages you to report any suspicious activity, including cybersecurity incidents, possible malicious code, software vulnerabilities, and phishing-related scams. Reporting forms can be found on CISA’s homepage at


  • August 26, 2020: Initial Version

This product is provided subject to this Notification and this Privacy & Use policy.

This post appeared first on
Author: CISA