@pisell/pisellos@2.3.139
一个可扩展的前端模块化SDK框架,支持插件系统
AI Security Review
scanned 1mo ago · by lpm-firewall-aiScanner labels are noisy: the critical Feishu posts are a named ScanOrder logger in a POS SDK, window eval is a timer polyfill, and RegisterAndLogin utils is field validation. With no install hook, no stealth harvest, and an export-only entrypoint, the package is clean.
Decision evidence
public snapshot- package.json has no preinstall, install, or postinstall hook; the only lifecycle script is prepublishOnly build.
- ScanOrder registers scanOrderLogger by default and points it at a hardcoded Feishu bot webhook when that solution initializes.
- window.js eval is a setTimeout string-handler fallback in a window polyfill, not hidden payload execution.
- RegisterAndLogin utils is ordinary email, phone, and password validation, not a harvest or exfil path.
- The published entrypoint only re-exports the PisellOS frontend SDK; import does not start network or install-time work.
- Feishu posting is a named ScanOrderLogger used after a host app starts ScanOrder or VenueBooking, which matches first-party POS debug logging.
- The HTTP plugin is a caller-driven request helper with no hardcoded attacker host.
- No child_process, filesystem writes, self-dependency, agent-config mutation, or reviewer-manipulation text was found.
Source & flagged code
5 flagged · loading sourceSource file is highly similar to a previously finalized malicious package; route for source-aware review.
dist/plugins/window.jsView on unpkgPackage source references a known benign dynamic code generation pattern.
dist/plugins/window.jsView on unpkg · L248Findings
Affected versions and remediation
This report applies to @pisell/pisellos@2.3.139.
See all versions of @pisell/pisellos for other recorded verdicts.
- Review the evidence and your use of @pisell/pisellos@2.3.139 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.