ClawHub
Back to skill

Security audit

Health Recorts

Security checks across malware telemetry and agentic risk

Overview

This health-record organizer handles sensitive medical files, but its instructions clearly require local processing, consent, and user-chosen storage boundaries.

Install this only if you want an agent to help organize medical records. Choose a private local folder, avoid cloud-synced storage unless intentional, decide retention/deletion expectations first, and approve any external OCR or API use only for specific files and trusted services.

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
92% confidence
Finding
The example triggers are very broad, everyday phrases that could activate the skill for loosely related requests without clearly scoping consent, data sensitivity, or intended handling. In a health-record ingestion skill, this is more dangerous because users may be led into sharing highly sensitive medical information before privacy expectations and boundaries are established.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The examples encourage building a reusable health workspace around PDFs, screenshots, scans, reports, and tracking data, but they provide no warning about handling sensitive medical information, retention, sharing, or access controls. In this context, the omission is especially risky because health data is highly sensitive and users may infer that bulk ingestion and long-term storage are routine and safe without being informed of privacy implications.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.