ClawHub
Back to skill

Security audit

unisound-med-clinical-decision

Security checks across malware telemetry and agentic risk

Overview

This skill is a transparent command-line clinical-assistance wrapper that reads user-supplied questions, optionally writes results, and sends the prompt to a disclosed medical model API.

Install only if you are comfortable sending the provided clinical question text to the configured medical-model API. De-identify patient data first, follow your organization’s privacy rules, and avoid passing real API keys in command histories or shared job logs when possible.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The skill documentation describes capabilities to read local input files, write output files, and send prompts to a remote medical-model API, but it does not declare corresponding permissions. This creates a transparency and governance gap: operators may deploy the skill without realizing it can access filesystem data and transmit potentially sensitive medical content over the network.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The code sends the full user prompt, which in this clinical-decision context may contain protected health information, to an external remote API endpoint without any built-in disclosure, consent check, minimization, or redaction. In a medical skill, this materially increases privacy and compliance risk because sensitive patient data can be transferred off-system or to an unintended processor.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
Accepting the API key via a command-line argument exposes it to shell history, process listings, job logs, and orchestration telemetry on many systems. While this is a common implementation shortcut, it is unsafe for secrets handling and can lead to credential compromise if the environment is multi-user or monitored.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.