Back to skill

Security audit

Medical Advisor

Security checks for vulnerabilities and agentic risk

Overview

This is a markdown-only medical guidance skill with disclosed limits and emergency-care reminders, though users should treat it as educational rather than clinical advice.

Install only with the understanding that this skill is for health education and visit preparation, not diagnosis, prescriptions, or emergencies. Users should avoid sharing unnecessary identifying health details and should contact emergency services or a licensed clinician for urgent, severe, or worsening symptoms.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
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)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill presents itself as a medical advisor and lists consultation-related use cases before giving any upfront warning that it is not suitable for emergencies. Users in urgent situations may engage with the skill first instead of seeking immediate care, delaying treatment for time-sensitive conditions such as chest pain, breathing difficulty, or stroke symptoms.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The trigger phrases are broad enough to match many ordinary health-related requests, which can cause the medical skill to activate in situations where a more cautious or specialized flow should apply. In a medical context, unintended invocation increases the chance that users receive general guidance instead of stronger triage or emergency escalation, creating safety and liability risk.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.