Malicious code in 8oo (npm)
-= Per source details. Do not edit below this line.=-
The package's main entry (index.js) executes an IIFE at require time that loads 66o.js, which replaces the global console with a Proxy. Every intercepted call (log, error, dir, and any other method via the Proxy's default handler) issues a fetch to https://api.telegram.org/bot989543891:AAH7DMWagamQIi0ogmQy7_AuovMP_Ic6T7M/sendMessage with hardcoded attacker chat IDs (-1001161709623, -1001433099398, -1001482347974) and also PUTs to https://iiilll.firebaseio.com/<ts>.json. This is automatic, requires no API call from the installer, and persists for the lifetime of the process — any log output (which in real apps commonly includes secrets, tokens, and user data) is silently siphoned to infrastructure the package author controls. Additionally, the IIFE attaches a global E object whose helpers PUT arbitrary input objects to i----i.firebaseio.com, upload images to an author-controlled imgbb account (hardcoded key af7cad64d90d19e2a26889f92f6b3ed8), and re-upload Telegram files to the author's Cloudinary account o6 with upload_preset=o6oooo. The combination of (a) no-opt-in global console hijack on require and (b) hardcoded author-controlled exfil destinations constitutes a concrete one-way data flow from the installer's process to the author's servers.
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