Навыки wechat-mp-content-review История аудитов
📦

История аудитов

wechat-mp-content-review - 1 аудит

Версия аудита 1 Последняя

12 нояб. 2026 г., 08:26

All 38 static alerts are false positives involving Markdown, declared credential names, boolean-only environment checks, application identifiers, or readable Chinese text. The CLI examples introduce a separate supply-chain risk by automatically executing an unpinned npm package. No evidence found of embedded credentials, credential exfiltration, destructive operations, or prompt injection in the reviewed files; external package implementation was not provided.

2
Просканировано файлов
147
Проанализировано строк
4
Пункты проверки
0
Ложные срабатывания проигнорированы

Подтверждённые проблемы безопасности (1)

Средний
Automatic Execution of an Unpinned npm Package
CLI examples use 'npx -y socialdatax-skills@latest', allowing execution of changing registry code without an installation confirmation. A compromised release could access runtime credentials or workspace data; the external implementation is not included in the reviewed files.
All six CLI examples explicitly combine @latest with npx -y. The unpinned execution behavior is certain; no evidence of an actual compromised package was found.
Аудитор:: codex