Security audit
Security checks for vulnerabilities and agentic risk
This is a disclosed market-strategy roleplay skill with broad activation terms but no hidden code, persistence, credential access, or destructive behavior.
Install this if you want a forceful market-strategy persona. Be aware it may activate on common market or competition terms and may give high-confidence, aggressive business advice, so users should ask it to exit role or request normal analysis when they need balanced risk review.
The trigger list includes very common words such as “市场”, “竞争”, and “对手”, which are likely to appear in ordinary conversation and can activate the skill without clear user intent. Because the skill then forces a strong persona and discourages meta-analysis, accidental activation can meaningfully steer responses toward aggressive, role-bound advice outside the user’s expectation.
The skill instructs the agent to immediately respond in-character, use a fixed persona voice, and avoid stepping out of role unless the user explicitly exits. This reduces user control and can mislead users into receiving stylized, high-certainty advice without clear ongoing consent, especially when combined with broad triggers that may activate the role unexpectedly.
No suspicious patterns detected.
