Security audit
Security checks for vulnerabilities and agentic risk
This is a persona-style strategy advice skill with broad activation wording, but it contains no code execution, persistence, credential access, or hidden data handling.
Install only if you want a Chinese historical-persona style for strategy and management advice. Treat investment, legal, medical, or other high-stakes answers as framing and brainstorming, not professional advice or an automatic decision-maker.
The skill mandates a fixed historical persona, tone, and output structure without checking user preference for language, style, or neutrality. While not directly enabling code execution or data exfiltration, this can degrade user autonomy, cause inappropriate stylistic responses in sensitive contexts, and increase the risk that advice is framed with undue authority.
The trigger condition is extremely broad, covering 'any' scenario involving strategy, decisions, management, investment, or competition. This can cause the skill to activate unintentionally in many ordinary conversations and override more appropriate system behavior, increasing the chance of misleading advice, role-confusion, or unsafe application in sensitive domains.
No suspicious patterns detected.
