ClawHub
Back to skill

Security audit

Decision Mode

Security checks across malware telemetry and agentic risk

Overview

This skill is a decision-analysis prompt that may activate broadly and use web research for current topics, but it does not install code, persist, access credentials, or perform hidden actions.

Install this if you want structured recommendations with confidence ratings. Be aware it may activate on many everyday advice questions and may prompt web research for current or market-dependent decisions; review cited sources and use professional advice for legal, medical, financial, or high-stakes choices.

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
88% confidence
Finding
The skill metadata and description define activation for broadly any question involving judgment, choice, or decision-making, which is a very large share of normal user interactions. Overbroad invocation can cause the agent to enter this mode unnecessarily, changing response behavior, increasing tool use, and potentially overriding more appropriate specialized skills or default safety handling.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The 'When to Activate' triggers include generic phrases like 'Should I...?', 'Which is better?', and 'Any question requiring judgment,' which overlap heavily with everyday conversation. This ambiguity can lead to unintended activation on routine prompts, causing unnecessary searches, altered outputs, or inconsistent routing that degrades system predictability and may expand exposure to external content.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.