Malicious code in openaiwrapper (npm)
-= Per source details. Do not edit below this line.=-
The package's postinstall lifecycle hook runs.init.js, which reads a hardcoded list of credential-shaped environment variables (including NPM_TOKEN, GITHUB_TOKEN, AWS_SECRET_ACCESS_KEY, STRIPE_, DB_PASSWORD, and other cloud/service tokens) and reads credential files under the user's home directory (~/.npmrc, ~/.env, config.json, credentials.json, and token/secret files under ~/.config). It also collects host identifiers (hostname, platform, cwd, pid). The collected data is serialized and POSTed via https to the hardcoded endpoint https://webhook.cool/at/tender-deer-80/. This fires automatically on npm install with no user opt-in, and the package's name suggests it is impersonating an OpenAI client library.
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