Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed markdown workflow for estimating occupational AI exposure and does not request dangerous access or perform hidden actions.
Before installing, be aware that the skill may activate on simple profession mentions and its numeric outputs are methodology-based estimates, not guaranteed source data. It does not appear to require sensitive permissions or local system access.
The skill activates on very broad signals such as any job title or generic mentions of AI replacement risk, which can cause unintended invocation during ordinary conversation. Over-broad triggering is risky because it may hijack unrelated user interactions, produce unsolicited analysis, and reduce the reliability and predictability of agent behavior.
The activation section explicitly lists broad triggers without constraints, so the skill may fire on incidental mentions of professions or AI-related topics rather than user intent to invoke this workflow. In context, this is not code-execution dangerous, but it can still cause scope overreach, incorrect routing, and unwanted responses that interfere with safer or more appropriate skills.
The manifest description includes both English and Chinese text, implying a language presentation choice has been made without stating that users can select their preferred language. Under the language/locale policy rule, skills should offer an explicit language choice or document why a fixed locale is required.
No suspicious patterns detected.
