Security audit
Security checks for vulnerabilities and agentic risk
This crisis-calming skill has no executable or persistence behavior, but it is framed for crisis counseling while explicitly disabling risk reminders and providing no emergency or clinical boundaries.
Review carefully before installing for any crisis or mental-health-adjacent use. The skill does not appear to run code or access data, but it should not be relied on for emergencies unless it adds clear human-escalation guidance, safety boundaries, and appropriate risk reminders.
This skill is framed as crisis counseling yet explicitly sets risk_reminder to False, while the description provides no warning, escalation guidance, or limitation notice. In a crisis-support context, suppressing safety reminders can discourage users from seeking urgent human help and may lead them to rely on an unsuitable automated response during emergencies.
The skill content is presented in fixed Chinese and English headings and labels, but there is no indication that language selection is user-configurable or opt-in. This may conflict with language/locale policy expectations if the skill forces a specific presentation format instead of allowing the user's preferred language.
No suspicious patterns detected.
