Security audit
Security checks for vulnerabilities and agentic risk
This skill is a coherent UIBC government-domain judging aid with only local criteria-checking behavior and no evidence of hidden access, persistence, or exfiltration.
Reasonable to install for UIBC GOV-domain judging or TH-GOV-001 criteria checks. Be aware that broad Chinese trigger terms such as government, migration judging, evidence package, and accountability may cause the skill to be selected in conversations that are not actually about the UIBC competition.
The example output states in Chinese that the criteria sections and response discipline must be quoted and not rewritten, which indicates the skill expects a fixed language/formulation. The README does not provide an opt-in language choice or justify a Chinese-only requirement as a locale-specific constraint.
The skill metadata includes broad trigger terms such as 'UIBC', '政务', '迁移判定', and '责任归属', which are common in ordinary discussion and not narrowly tied to this specific skill. In an agent routing system, this can cause unintended activation, pulling the model into using domain-specific rules or accountability language in contexts where the user did not request this skill, increasing the risk of misrouting and confusing or overly authoritative responses.
The usage section says to use the skill whenever the user mentions broad topical keywords and to treat the package's criterion file as the required source, but it does not define boundaries or exclusions. This ambiguity makes accidental invocation more likely and can bias outputs toward this skill's framing, which is especially risky because the skill claims judging and accountability authority that may not fit the user's actual task.
