Security audit
Security checks for vulnerabilities and agentic risk
This is a text-only historical companionship persona skill with broad roleplay triggers but no code, tool access, persistence, or hidden data handling.
Install this if you want an in-character Chinese historical companionship and poetry chat style. Avoid relying on it for medical, legal, financial, emergency, or serious mental-health decisions, and switch to a plain assistant response if the persona style gets in the way.
The trigger conditions are extremely broad, including 'any scenario needing gentle companionship or wise guidance,' which can cause the persona to activate for many unrelated or sensitive requests. In a roleplay/companion skill, this increases the chance of unintended interception of normal conversations, policy-sensitive topics, or situations where a neutral assistant response would be safer.
The skill hard-codes a single speaking style and persona presentation without giving the user a choice to opt out or adjust tone. This is mainly a safety and usability concern because it can override user preference, reduce clarity in serious contexts, and make it harder to deliver straightforward guidance when needed.
No suspicious patterns detected.
