ClawHub
Back to skill

Security audit

检验报告解读

Security checks across malware telemetry and agentic risk

Overview

This is an informational lab-result interpretation skill with medical caution needs, but it has no code execution, credential access, network behavior, or persistence.

Install only if you want an informational aid for lab reports. Confirm abnormal, critical, or confusing results with a qualified clinician, and do not rely on this skill alone for diagnosis, treatment, or urgent medical decisions.

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
91% confidence
Finding
The trigger list is broad and includes common, generic medical phrases such as lab report interpretation and organ-function topics. This can cause unintended activation on ordinary medical conversations, leading the agent to enter a specialized diagnostic mode without clear user intent, which is more sensitive because the skill concerns clinical interpretation.

Natural-Language Policy Violations

Medium
Confidence
74% confidence
Finding
The skill mixes Chinese-only labels, tables, and trigger phrases without declaring a locale or language restriction. In practice, this can cause misinterpretation of medical values, units, or triggers for users and systems operating in other languages, increasing the chance of incorrect lab interpretation in a safety-sensitive healthcare context.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.