Security audit
Security checks for vulnerabilities and agentic risk
This is a Chinese writing-assistance skill for rural education volunteering, with no code execution, data access, persistence, or hidden authority.
Installers should expect this skill to shape responses toward warm Chinese-language rural education volunteer writing. If broad trigger activation is undesirable, the publisher could narrow the trigger terms, but no security-sensitive behavior was found.
The trigger list includes broad terms such as '寄语' and '志愿者' that can appear in many benign conversations outside the intended rural-education context. This can cause unintended activation or routing, leading to prompt hijacking of normal conversations into the skill’s persona or accidental disclosure of skill-specific behavior.
The description mandates Chinese-language behavior by default, which can override a user’s preferred language or system-level localization expectations. While not directly a security exploit, forced language behavior can degrade user control, create misleading outputs, and interfere with safe handling if users expect responses in another language.
No suspicious patterns detected.
