神經科病歷助手

Security checks across malware telemetry and agentic risk

Overview

This is a simple neurology note-structuring skill, but users should handle patient records carefully and have a qualified clinician review any medical suggestions.

Install only if you are authorized to process the medical text you provide. De-identify patient records where possible, use an approved privacy environment, and treat diagnosis or treatment suggestions as drafting support that must be reviewed by a qualified clinician before clinical use.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill is explicitly designed to process outpatient medical records, which are highly sensitive personal and health data, but it provides no warning about privacy, consent, retention, or de-identification requirements. In a medical context, this omission materially increases the risk that users will paste protected health information into an AI workflow without appropriate safeguards, creating confidentiality, compliance, and patient-safety exposure.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal