Security audit
Security checks for vulnerabilities and agentic risk
This skill is a straightforward medical-document FAQ generator with some privacy and scope caveats but no evidence of hidden, destructive, persistent, or exfiltrating behavior.
Before installing, treat any source medical document as potentially sensitive. Use sanitized or minimally necessary documents, avoid PHI unless your environment is approved for it, and review generated FAQs for medical accuracy and privacy before sharing them.
The trigger description is broad enough to match common medical explanation or patient-education requests beyond narrowly scoped FAQ generation. This can cause the skill to activate in situations involving sensitive medical content or advice-like requests without clear boundaries, increasing the chance of inappropriate handling, oversharing, or user confusion about the tool's role.
The skill processes medical documents, which often contain PHI, confidential policy details, or other sensitive data, yet the description provides no warning or handling guidance for such content. Without explicit safeguards, users or orchestrators may pass sensitive records into the skill and receive outputs that reproduce or expose private information.
No suspicious patterns detected.
