Security audit
Security checks for vulnerabilities and agentic risk
This is a conversational restaurant food-cost diagnostic skill with disclosed, purpose-aligned tracking and no executable or privileged behavior.
Before installing, understand that the skill may retain food-cost percentages, targets, root causes, actions, and follow-up dates for pattern tracking. Use explicit commands when you want a diagnostic, and avoid entering sensitive business details you do not want stored in agent memory.
The skill advertises activation by saying "food cost is high," which is a natural phrase an operator might use casually in discussion rather than as an intentional invocation. Because the document does not provide tighter trigger boundaries or negative examples, this could cause unintended activation during ordinary conversation about restaurant performance.
The trigger section says to run when food cost "feels off" or when pattern tracking shows an issue due for follow-up, but it does not clearly distinguish discussion signals from actual run commands. This ambiguity makes it unclear when the skill should activate versus when it should simply continue normal conversation.
No suspicious patterns detected.
