ClawHub
Back to skill

Security audit

healthcare-dev

Security checks across malware telemetry and agentic risk

Overview

This is a healthcare software guidance skill with sensitive example code, but the artifacts are instruction-only and the risky behaviors are disclosed and purpose-aligned.

Install this only for healthcare software engineering guidance. Treat the examples as starting points, not production-ready healthcare code; validate PHI logging, credential handling, encryption, retention, access controls, telemedicine recording consent, and jurisdiction-specific compliance with qualified security, privacy, legal, and clinical reviewers.

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 example enables consultation recording based solely on an environment flag and does not enforce explicit patient/provider consent, notice, or jurisdiction-specific recording requirements before recording starts. In a telemedicine context, this can lead to unauthorized capture of highly sensitive health information, creating significant privacy, legal, and compliance exposure.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal