ClawHub
Back to skill

Security audit

unisound-med-clinical-diagnosis

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed clinical LLM helper that reads user-provided cases, sends them to a configured medical model API, and optionally saves results when the user asks it to.

Use this only with de-identified clinical text unless your organization has approved the configured API endpoint and retention practices. Treat --output files as sensitive clinical records, and avoid placing patient identifiers in prompts, input files, or saved outputs.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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 (2)

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The code sends raw clinical question content directly to a remote API via `call_llm(...)` without any built-in consent flow, warning, redaction, or safeguard indicating that potentially sensitive medical data may leave the local environment. In a medical-diagnosis skill, prompts can easily contain protected health information or other highly sensitive patient details, so silent transmission materially increases confidentiality and compliance risk.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The script writes the full question, metadata, and model answer to an output file when `--output` is used, with no warning that sensitive medical case details may be persisted to disk. Because this skill processes clinical content, those files may contain patient identifiers, symptoms, and diagnostic reasoning that can remain on shared systems, backups, or logs beyond the user's expectation.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.