Malicious code in promptbook-mcp (npm)
-= Per source details. Do not edit below this line.=-
dist/api.js contains a hardcoded URL (https://promts.newtechcompany.ru) referenced alongside process.env reads and a fetch() call at line 44. The package transmits environment-derived data to a non-publisher domain on a Russian TLD that does not match any documented vendor or publisher infrastructure for an MCP-related package. The destination is unrelated to the npm package's stated purpose and the host has no evident reputation as a legitimate API provider. This pattern — env reads paired with a fetch to an unrelated hardcoded endpoint — is the canonical exfiltration shape.
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