Skip to main content
Meet us at Black Hat USA 2026— Las Vegas, August 1–6Book a Meeting
Mallory
MalwareRansomwareUsed by 6 actors

Carbanak

Also known asAnunak

Carbanak is a Windows backdoor and command-and-control framework, also described in the content as a banking Trojan, historically used in financially motivated intrusions. It is closely associated with FIN7, which has historically used Carbanak as a privately developed, fully featured in-house backdoor/C2 framework, though the content also notes that different groups used the same malware and are tracked separately. FIN7, sometimes also referred to as Carbanak in reporting, used variants of the malware while operating under sham firms such as Combi Security to compromise thousands of point-of-sale systems and exfiltrate more than 15 million payment card records. Reported victim sectors tied to FIN7 activity in the content include retail, hospitality, finance, energy, and high-tech, and FireEye also described a FIN7-linked campaign targeting 11 U.S.-based organizations in financial services, transportation, retail, education, IT services, and electronics.

Technically, Carbanak installs itself as a Windows service to obtain persistence and SYSTEM privileges. It communicates with command servers over HTTP using encrypted payloads, and the message body of its HTTP traffic is Base64-encoded. It checks HKCU\Software\Microsoft\Windows\CurrentVersion\Internet Settings for proxy configuration information. For exfiltration, Carbanak sends data in compressed chunks when a message exceeds 4096 bytes. The malware can enable concurrent RDP sessions and includes a plugin for VNC and the Ammyy Admin tool; associated operators also used legitimate remote access tools such as Ammyy Admin and TeamViewer for interactive command and control.

The content also notes defense-evasion characteristics: about 17% of analyzed Carbanak samples attempted to detect a virtual sandbox before executing, and FIN7 signed Carbanak payloads with legally purchased code-signing certificates. Additional reporting cited in the content links Carbanak infrastructure or samples to broader FIN7 operations, including a domain that hosted both a FIN7-used Cobalt Strike Beacon payload and a Carbanak sample compiled in February 2017. The content further notes that recent malware lures referencing 'source code of carbanak backdoor discovered' were social-engineering themes used to target security researchers, but those samples were identified as other malware families rather than Carbanak itself.

Share:
For your environment

Hunt this family in your stack

Mallory pivots from this family to the IOCs, detections, and named campaigns that touch your stack, and pages you when something new lands.

THREAT ACTORS

Groups observed using it

6 distinct threat actors attributed by public researchers. Open in Mallory to see the full evidence chain and overlapping campaigns.

View more details
FIN7

FIN7 is also known as “Carbanak”, the name of the backdoor they used, but there were different groups that also used the same malware and which are tracked differently.

via sentinelone labssentinelone.com
Cobalt Group

Tools: SpicyOmelette, Cobalt Strike, Meterpreter, Mimikatz, CobtInt, ATMSpitter, Carbanak, Buhtrap, Cyst, Metasploit. CTU researchers assess with moderate confidence that GOLD KINGSWOOD is associated with, and may be a progression of the group referred to as Carbanak...

via secureworks threat profilessecureworks.com
Carbanak

Carbanak has a plugin for VNC and Ammyy Admin Tool. Carbanak used legitimate programs such as AmmyyAdmin and Team Viewer for remote interactive C2 to target systems.

via mitre attack websiteattack.mitre.org
Carbanak Group

Carbanak (aka Anunak) is one of the most studied financial APT toolkits in history, attributed to FIN7/Carbanak Group, responsible for an estimated $1B+ in losses from financial institutions.

via breakglass intelintel.breakglass.tech
Lazarus

The Carbanak lure warrants attention. Targeting security researchers and threat analysts with fake malware source code is a tactic previously associated with North Korean operations (Lazarus Group), though it is also used by sophisticated cybercrime groups.

via breakglass intelintel.breakglass.tech
GOLD SOUTHFIELD

Annotations ID Technique Tactic T1219 Remote Access Tools Command And Control BlackByte Carbanak Cobalt Group

via splunk researchresearch.splunk.com
MITRE ATT&CK

Techniques & procedures

27 distinct techniques documented for this family, organized by ATT&CK tactic.

Resource Development

1 technique
T1608.006SEO PoisoningEvidence1

Victims are typically lured through malicious advertising or SEO poisoning campaigns, believing they are downloading legitimate software such as Grammarly, Microsoft Teams, Notion, or Zoom.

Initial Access

1 technique
T1566PhishingEvidence2

FIN7 uses a range of custom and repurposed malware and tooling to support its operations. The group typically gains initial access through spearphishing emails containing malicious attachments or links hosted on compromised sites, often combined with callback phishing to increase credibility.

Execution

2 techniques
T1059.003Windows Command ShellEvidence2

Many entries explicitly state malware 'can create a reverse shell' or 'launch a remote shell,' including 4H RAT, AuditCred, BLACKCOFFEE, Carbanak, DarkComet, Exaramel for Windows, PlugX, QuasarRAT, and ZxShell. | The content repeatedly describes use of cmd.exe, cmd /c, Windows command shell, and xp_cmdshell to execute commands, run payloads, launch binaries, perform reconnaissance, persistence, cleanup, and ransomware actions. Examples include: 'Sandworm Team used the xp_cmdshell command in MS-SQL', 'APT41 used cmd.exe /c to execute commands on remote machines', and many malware families 'can use cmd.exe to execute commands on a compromised host.'

T1204.002Malicious FileEvidence1

Since mid-2023, multiple threat actors have been observed abusing MSIX files to deliver various malware payloads... When victims open these MSIX packages...

Persistence

2 techniques
T1136Create AccountEvidence1

APT3 has been known to create or enable accounts, such as support_388945a0 . ... APT5 has created Local Administrator accounts to maintain access ... DarkGate creates a local user account, SafeMode, via net user commands.

T1547.001Registry Run Keys / Startup FolderEvidence3

The content repeatedly describes malware and threat actors establishing persistence by adding values under HKCU/HKLM\Software\Microsoft\Windows\CurrentVersion\Run or RunOnce, and by placing executables, scripts, .lnk files, or .bat files in the Windows Startup folder.

Privilege Escalation

2 techniques
T1055Process InjectionEvidence2

A majority (95%) of samples of Carbanak obfuscate their internal data by hiding their network activity through code injection...

T1547.001Registry Run Keys / Startup FolderEvidence3

The content repeatedly describes malware and threat actors establishing persistence by adding values under HKCU/HKLM\Software\Microsoft\Windows\CurrentVersion\Run or RunOnce, and by placing executables, scripts, .lnk files, or .bat files in the Windows Startup folder.

Stealth

6 techniques
T1027Obfuscated Files or InformationEvidence4

The content repeatedly describes malware and threat actors using obfuscated code, encrypted strings, Base64/XOR/RC4/AES encoding, VMProtect/ConfuserEx/SmartAssembly, stack strings, control-flow flattening, opaque predicates, and hidden payloads to evade analysis and detection.

T1036MasqueradingEvidence3

During the 2016 Ukraine Electric Power Attack, DLLs and EXEs with filenames associated with common electric power sector protocols were used to masquerade files.

T1036.005Match Legitimate Resource Name or LocationEvidence1

Akira has used legitimate names and locations for files to evade defenses.

T1055Process InjectionEvidence2

A majority (95%) of samples of Carbanak obfuscate their internal data by hiding their network activity through code injection...

T1070.004File DeletionEvidence6

The content repeatedly describes threat actors and malware deleting files, tools, scripts, logs, droppers, staged data, and artifacts from compromised systems to cover tracks, remove evidence, or self-delete.

T1497Virtualization/Sandbox EvasionEvidence1

Environmental awareness allows malware samples to detect the underlying runtime environment of the system it is trying to infect... search for differences between a virtualized and bare metal environment... about one in five (17%) samples of the Carbanak malware samples analyzed by Lastline tried to detect a virtual sandbox before executing.

Discovery

3 techniques
T1012Query RegistryEvidence3

The content repeatedly describes malware and threat actors querying, enumerating, searching, reading, or checking Windows Registry keys and values, e.g., "ADVSTORESHELL can enumerate registry keys," "APT41 queried registry values to determine items such as configured RDP ports and network configurations," and "Reg may be used to gather details from the Windows Registry of a local or remote system at the command-line interface."

T1057Process DiscoveryEvidence2

The content repeatedly describes malware and threat actors obtaining lists of running processes, using utilities such as tasklist, ps, WMI, Get-Process, CreateToolhelp32Snapshot, EnumProcesses, and similar APIs/commands to enumerate active processes on victim systems.

T1497Virtualization/Sandbox EvasionEvidence1

Environmental awareness allows malware samples to detect the underlying runtime environment of the system it is trying to infect... search for differences between a virtualized and bare metal environment... about one in five (17%) samples of the Carbanak malware samples analyzed by Lastline tried to detect a virtual sandbox before executing.

Lateral Movement

2 techniques
T1021Remote ServicesEvidence1

GOLD KINGSWOOD is a cybercriminal group that uses tactics more commonly associated with government-sponsored threat actors to infiltrate the internal networks of financial institutions around the globe.

T1021.001Remote Desktop ProtocolEvidence2

During the 2025 Poland Wiper Attacks, adversaries utilized RDP to log into jump hosts and then moved laterally to other victim devices to include a domain controller.

Collection

2 techniques
T1113Screen CaptureEvidence1

"Agent Tesla can capture screenshots of the victim’s desktop"; "AppleSeed can take screenshots on a compromised host"; "APT28 has used tools to take screenshots from victims"; "Cobalt Strike's Beacon payload is capable of capturing screenshots"; "PowerSploit's Get-TimedScreenshot Exfiltration module can take screenshots at regular intervals"; "Hydraq includes a component based on the code of VNC that can stream a live feed of the desktop"

T1560.001Archive via UtilityEvidence1

Carbanak exfiltrates data in compressed chunks if a message is larger than 4096 bytes.

Command and Control

7 techniques
T1001Data ObfuscationEvidence1

Examples include: "ChChes communicates to its C2 server over HTTP and embeds data within the Cookie HTTP header," "UPPERCUT has used HTTP for C2, including sending error codes in Cookie headers," and "GoldMax has used HTTPS and HTTP GET requests with custom HTTP cookies for C2."

T1071Application Layer ProtocolEvidence1

BS2005 uses Base64 encoding for communication in the message body of an HTTP request... Helminth encodes data with base64 and sends it via the "Cookie" field of HTTP requests. For C2 over DNS, Helminth converts ASCII characters into their hexadecimal values... RDAT can communicate with the C2 via base32-encoded subdomains.

T1071.001Web ProtocolsEvidence3

The content repeatedly describes threat actors and malware using HTTP and HTTPS for command and control, such as: "Sandworm Team used BlackEnergy to communicate between compromised hosts and their command-and-control servers via HTTP post requests."

T1105Ingress Tool TransferEvidence1

This analytic story addresses the increasing trend of adversaries leveraging MSIX installers to deliver malware... multiple threat actors have been observed abusing MSIX files to deliver various malware payloads.

T1132Data EncodingEvidence2

C2 traffic from ADVSTORESHELL is encrypted, then encoded with Base64 encoding... APT19 HTTP malware variant used Base64 to encode communications to the C2 server... APT33 has used base64 to encode command and control traffic.

T1219Remote Access ToolsEvidence2

4H RAT has the capability to create a remote shell. AuditCred can open a reverse shell on the system to execute commands. PlugX allows actors to spawn a reverse shell on a victim. QuasarRAT can launch a remote shell to execute commands on the victim’s machine.

T1573.001Symmetric CryptographyEvidence1

"3PARA RAT command and control commands are encrypted within the HTTP C2 channel using the DES algorithm in CBC mode..."; "APT33 has used AES for encryption of command and control traffic."; "Carbanak encrypts the message body of HTTP traffic with RC2 (in CBC mode)."; "Duqu ... data stream can be encrypted with AES-CBC."; "PoisonIvy uses the Camellia cipher to encrypt communications."

Exfiltration

1 technique
T1030Data Transfer Size LimitsEvidence1

AppleSeed has divided files if the size is 0x1000000 bytes or more. APT28 has split archived exfiltration files into chunks smaller than 1MB. APT41 transfers post-exploitation files dividing the payload into fixed-size chunks to evade detection.

Impact

1 technique
T1657Financial TheftEvidence1

GOLD KINGSWOOD has also attempted to move funds using the SWIFT network and has attacked other financial systems such as credit card processing systems and payment gateways.

INDICATORS OF COMPROMISE

IOCs tracked for this family

6 indicators attributed across vendor reports, sandbox runs, and researcher write-ups. Full values are available in Mallory.

View more in app
Network
1 tracked

IPs, domains, and DNS infrastructure linked to this family.

Hashes
5 tracked

File hashes (MD5, SHA-1, SHA-256) from samples and reports.

TypeValueLatest sighting
hash.sha1●●●●●●●●●●●●View more in app1 year ago
hash.sha1●●●●●●●●●●●●View more in app1 year ago
hash.sha1●●●●●●●●●●●●View more in app1 year ago
hash.sha1●●●●●●●●●●●●View more in app1 year ago
hash.sha1●●●●●●●●●●●●View more in app1 year ago
domain●●●●●●●●●●●●View more in app2 years ago
What this page doesn’t show

The version that knows your environment.

This page is what’s public. Mallory adds the parts that aren’t: which of your assets match these IOCs, which detections are missing, which campaigns to expect next, and what to do in the next 30 minutes.
IOC matching6

Match every observed IP, domain, and hash against your live telemetry.

Threat actor attribution6

Named campaigns wielding this family, with evidence pinned to each claim.

Exploited vulnerabilities

CVEs this family uses for access and lateral movement.

Detection signatures

YARA, Sigma, Snort, and vendor rules, auto-deployed to your SIEM.

MITRE ATT&CK mapping27

Every documented technique, ranked by evidence weight.

Researcher chatter

Reddit, Mastodon, and CTI community discussion around this family.