ClawHub
Back to skill

Security audit

Logic Inversion Reflector

Security checks across malware telemetry and agentic risk

Overview

This is a local decision-reflection skill that challenges assumptions and shows no hidden data access, network use, persistence, or privileged behavior.

Safe to install from a security perspective. Use it as a thinking aid for non-urgent decisions, and be aware that it is intentionally confrontational; disable or ignore it when you do not want assumption-challenging feedback.

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
90% confidence
Finding
The skill allows automatic activation whenever it 'detects a major decision,' which is subjective and undefined. In practice this can cause unsolicited intervention, derail normal conversations, and override user intent by injecting adversarial-style challenge prompts into sensitive or time-critical contexts.

Vague Triggers

Low
Confidence
87% confidence
Finding
The instruction to use LIR 'naturally in conversation' is overly broad and lacks bounded conditions, making it easy for the skill to activate in unrelated discussions. While less severe than the auto-detection trigger, it still risks unwanted cognitive pressure, conversation hijacking, and inconsistent behavior relative to user expectations.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal