ClawHub
Back to skill

Security audit

unisound-med-rx-safety-ethics

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed command-line wrapper that sends user-provided medical questions to a specified medical LLM API, with no hidden persistence or unrelated behavior found.

Install only if you are comfortable sending question text to the configured external medical-model API. Do not include real patient identifiers or protected health information unless your organization has approved that workflow, and use --dry-run when you only want to verify parsing without a network call.

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 (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding
The skill documentation advertises capabilities that read local files and stdin, write output files, and send prompts to a remote medical-model API, but it declares no corresponding permissions. This creates a trust and policy gap: operators may deploy the skill without realizing it can exfiltrate sensitive medical questions or patient-related content over the network and persist data to disk.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill sends user-supplied medical questions to a remote third-party API, and those questions may contain sensitive health or personal data. There is no consent flow, warning, redaction step, or clear disclosure before transmission, which creates a real privacy and compliance risk in a medical context where data sensitivity is unusually high.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.