### Summary
A malware campaign is targeting AI developers and software engineers through trojanized GitHub repositories impersonating popular AI tools and developer resources. Attackers clone legitimate repositories, modify installation instructions or embed malicious payloads, and deliver a multi-stage SmartLoader infection chain that ultimately installs credential-stealing malware. The campaign leverages EtherHiding, resolving command-and-control (C2) infrastructure from the Polygon blockchain at runtime, enabling attackers to dynamically update C2 servers while evading traditional detection methods. The activity primarily targets organizations in the financial services, banking, and technology sectors.[/subscribe_to_unlock_form]
### Summary
A malware campaign is targeting AI developers and software engineers through trojanized GitHub repositories impersonating popular AI tools and developer resources. Attackers clone legitimate repositories, modify installation instructions or embed malicious payloads, and deliver a multi-stage SmartLoader infection chain that ultimately installs credential-stealing malware. The campaign leverages EtherHiding, resolving command-and-control (C2) infrastructure from the Polygon blockchain at runtime, enabling attackers to dynamically update C2 servers while evading traditional detection methods. The activity primarily targets organizations in the financial services, banking, and technology sectors.[emaillocker id="1283"]
The attack begins with fake repositories for tools such as Claude, ComfyUI, AI coding assistants, Python security guides, and Rust frameworks. Victims download a ZIP archive containing obfuscated Lua scripts executed through SmartLoader, which performs reconnaissance, geolocation checks, and blockchain-based C2 resolution before retrieving additional payloads. The final stage deploys NodeJS-based infostealers designed to steal browser credentials, cloud API keys, authentication tokens, and other sensitive developer assets.
Successful compromise of developer workstations could lead to credential theft, unauthorized access to cloud environments, CI/CD pipelines, and software repositories, creating opportunities for downstream supply chain attacks. Organizations should verify the authenticity of GitHub repositories before downloading developer tools, enforce least-privilege access, monitor for suspicious installation scripts, and rotate exposed credentials and API keys if compromise is suspected.
| Tactic | Technique Id | Technique | Sub-technique |
|---|---|---|---|
| Initial access | T1195 | Supply Chain Compromise | - |
| Initial access | T1566.003 | Phishing | Spearphishing via Service |
| Persistence | T1547.001 | Boot or Logon Autostart Execution | Registry Run Keys / Startup Folder |
| Defence Evasion | T1027.002 | Obfuscated Files or Information | Software Packing |
| Credential access | T1555.003 | Credentials from Password Stores | Credentials from Web Browsers |
| Command and control | T1071.001 | Application Layer Protocol | Web Protocols |
| Command and control | T1571 | Non | Standard Port- |
| Exfiltration | T1041 | Exfiltration Over C2 Channel | - |
| Objective | Behavior ID | Behavior |
|---|---|---|
| Defense Evasion | B0029 | Polymorphic Code |
| Anti-Static Analysis | B0032 | Executable Code Obfuscation |
| Command & Control | B0030 | C2 Communication |
| Impact | B0022 | Remote Access |
| Discovery | E1083 | File and Directory Discovery |
| Execution | E1204 | User Execution |
| Discovery | E1082 | System Information Discovery |
| Anti-Behavioral Analysis | B0003 | Dynamic Analysis Evasion |
| Exfiltration | E1020 | Automated Exfiltration |
The following reports contain further technical details:
[/emaillocker]