Security audit
Security checks for vulnerabilities and agentic risk
This is a text-only decision helper with broad trigger phrases but no evidence of hidden access, execution, persistence, or data handling.
Before installing, consider whether you want a general decision-routing framework to activate on broad option-selection questions. It appears safe for structured comparison, but it should not be relied on for regulated financial, legal, tax, fiduciary, or procurement approvals.
The trigger set as a whole is insufficiently constrained, containing multiple generic decision-making phrases without exclusions or domain qualifiers. This broad matching surface increases unintended activation risk, creates overlap with normal assistant behavior, and may let the skill influence conversations outside its stated role of structured route comparison.
The trigger phrase "best option" is extremely generic and is likely to match ordinary user requests far outside the intended routing/selection domain. This can cause the skill to activate unexpectedly, hijack unrelated conversations, and route user input into decision logic without clear user intent or scope boundaries.
The phrase "which path should I choose" is ambiguous and commonly used in everyday conversation, making accidental invocation likely. In a skill that provides strategic routing recommendations, broad activation increases the chance of unsolicited or contextually inappropriate advice being injected into unrelated user interactions.
The activation rule includes the phrase “which of these should I choose?”, which is broad natural language that commonly appears in ordinary conversation. This can cause unintended skill activation and routing behavior in contexts outside the intended decision-analysis use case, potentially overriding a more appropriate skill or response path.
No suspicious patterns detected.
