Back to skill

Security audit

Apple Health Tool Free

Security checks across malware telemetry and agentic risk

Overview

This skill is a plausible Apple Health assistant, but it handles sensitive health data through an external API with broad activation wording and inconsistent free-tier capability instructions.

Install only if you intend to use Transition with Apple Health data. Treat the API key like a password, avoid broad automatic use for generic analysis tasks, and review any local cache or shell-profile changes before allowing the agent to write them.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Description-Behavior Mismatch

High
Confidence
91% confidence
Finding
The manifest states PMC is not supported in the free tier, yet later documentation instructs users to call the PMC endpoint directly. This mismatch can cause an agent to invoke a privileged or unintended endpoint, potentially exposing additional health-derived performance data or circumventing product restrictions if backend enforcement is weak.

Vague Triggers

Medium
Confidence
82% confidence
Finding
The trigger conditions are broad enough that an agent may invoke this skill for generic analysis, reporting, or visualization tasks without sufficiently confirming that the user intends to access sensitive health data through an external service. In a health context, over-broad activation increases the chance of unnecessary transmission of private data and inappropriate tool execution.

Missing User Warnings

High
Confidence
97% confidence
Finding
The skill handles highly sensitive health information but does not prominently warn that queries may transmit Apple Health-derived data to the external Transition API and possibly through the agent runtime. Without explicit disclosure and consent language, users may unknowingly expose medical and fitness data outside their device.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.