醫療病歷結構化助手

Security checks across malware telemetry and agentic risk

Overview

This is a simple medical-note structuring prompt with no code or system access, but users should treat its clinical outputs as drafts requiring clinician review.

Install only if you intend to use it for medical-record summarization in an environment appropriate for patient data. Any diagnosis, treatment plan, or medication recommendation should be reviewed and approved by a qualified clinician before use.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
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

Findings (2)

Vague Triggers

Medium

Confidence: 89% confidence
Finding: The skill’s invocation scope is extremely broad ('any medical personnel needing quick summaries') and the example trigger is effectively any patient symptoms and exam results. In a medical context, vague activation criteria can cause the assistant to be used for unintended clinical tasks or on inappropriate inputs, increasing the chance of unsafe summarization, overreliance, or accidental handling of sensitive medical data without clear boundaries.

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The description states the skill will automatically structure outpatient/inpatient medical records and output diagnoses and treatment recommendations, but it provides no activation boundaries, user-role restrictions, safety checks, or clinical disclaimers. In a medical context, broad and underspecified behavior is dangerous because it can be invoked on sensitive health data and may generate clinical advice beyond an appropriate scope, increasing the risk of privacy violations and unsafe medical decision support.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal