LAMEHUG is Python-based information stealer first identified in July 2025 by Ukraine's Computer Emergency Response Team (CERT-UA) in phishing emails targeting Ukrainian government officials. LAMEHUG is the first known malware to integrate artificial intelligence (AI) directly into its attack workflow by querying large language models (LLMs) hosted on Hugging Face to dynamically generate reconnaissance, data theft, and system manipulation commands in real time. LAMEHUG has been attributed to APT28. [1][2][3]
| Name | Description |
|---|---|
| PROMPTSTEAL |
| Domain | ID | Name | Use | |
|---|---|---|---|---|
| Enterprise | T1087 | .002 | Account Discovery: Domain Account |
LAMEHUG can use dsquery to enumerate domain user information.[3] |
| Enterprise | T1071 | .001 | Application Layer Protocol: Web Protocols |
LAMEHUG can use HTTP POST requests to exfiltrate data from compromised hosts to C2.[1] |
| Enterprise | T1560 | .001 | Archive Collected Data: Archive via Utility |
LAMEHUG can xcopy for file collection on targeted systems.[1] |
| Enterprise | T1119 | Automated Collection |
LAMEHUG can recursively copy files from targeted directories on victim hosts.[1][2] |
|
| Enterprise | T1059 | .003 | Command and Scripting Interpreter: Windows Command Shell |
LAMEHUG can use |
| .006 | Command and Scripting Interpreter: Python | |||
| Enterprise | T1132 | Data Encoding | ||
| Enterprise | T1005 | Data from Local System |
LAMEHUG has the ability to collect system information and files of interest from compromised systems.[1][2] |
|
| Enterprise | T1074 | .001 | Data Staged: Local Data Staging |
LAMEHUG can save collected data and files of interest in |
| Enterprise | T1140 | Deobfuscate/Decode Files or Information |
LAMEHUG can decode and drop a decoy file attached to spearphishing emails.[1] |
|
| Enterprise | T1482 | Domain Trust Discovery | ||
| Enterprise | T1573 | .002 | Encrypted Channel: Asymmetric Cryptography | |
| Enterprise | T1041 | Exfiltration Over C2 Channel |
LAMEHUG can exfiltrate collected system information and documents to C2.[1][2] |
|
| Enterprise | T1083 | File and Directory Discovery |
LAMEHUG can target directories on victim machines for file collection.[1][2] |
|
| Enterprise | T1036 | .005 | Masquerading: Match Legitimate Resource Name or Location |
LAMEHUG payloads have been disguised with legitimate looking filenames including AI_generator_uncensored_Canvas_PRO_v0.9.exe and AI_image_generator_v0.95.exe.[1][2] |
| Enterprise | T1069 | .002 | Permission Groups Discovery: Domain Groups |
LAMEHUG can use dsquery to gather domain group information.[3] |
| Enterprise | T1566 | .001 | Phishing: Spearphishing Attachment |
LAMEHUG has been distributed through spearphishing emails with various AI-themed malicious attachments.[1] |
| Enterprise | T1057 | Process Discovery |
LAMEHUG can gather process information on targeted systems.[2][3] |
|
| Enterprise | T1082 | System Information Discovery |
LAMEHUG has the ability to execute Windows commands returned from C2 to gather system information.[1][2] |
|
| Enterprise | T1016 | System Network Configuration Discovery |
LAMEHUG can enumerate network information on compromised hosts.[2] |
|
| Enterprise | T1033 | System Owner/User Discovery | ||
| Enterprise | T1007 | System Service Discovery |
LAMEHUG can gather service information on targeted systems.[2][3] |
|
| Enterprise | T1204 | .002 | User Execution: Malicious File |
LAMEHUG has been executed through victim interaction with malicious email attachments made to look like legitimate AI applications or documents.[1] |
| Enterprise | T1102 | .002 | Web Service: Bidirectional Communication |
LAMEHUG has used the Hugging Face API to query the Qwen2.5-Coder-32B-Instruct LLM to generate one-line Windows commands for the collection of system information and documents in specific folders on compromised hosts. LAMEHUG subsequently executed the returned commands and exfiltrated the collected files and information to adversary-controlled C2 servers.[2][1] |
| Enterprise | T1047 | Windows Management Instrumentation | ||