ClawHub
Back to skill

Security audit

ai-consciousness

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed AI behavior and governance prompt skill with broad activation, but it contains no executable code, hidden data access, or destructive behavior.

Install this only if you want a broad, always-available AI behavior/governance framework. It is not showing malicious behavior, but its generic triggers may make it appear in more conversations than expected, so users who want narrow task-specific skills may prefer tighter activation wording.

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 (3)

Vague Triggers

Medium
Confidence
89% confidence
Finding
The trigger keywords include extremely broad terms such as "ai" and "consciousness", which can cause this skill to activate in many unrelated conversations. Because the skill is configured to shape agent behavior and governance framing, unintended activation can alter responses outside its intended scope and create prompt-scope interference with other skills or user instructions.

Vague Triggers

Medium
Confidence
86% confidence
Finding
The read_when conditions are very broad, including "every session start," "before delivering outputs," and when the user expresses complaints or errors occur. This effectively makes the skill omnipresent, increasing the chance that its normative instructions influence unrelated tasks and override more context-appropriate behavior, especially since it is an AI behavior framework rather than a narrowly scoped utility.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The manifest advertises many generic trigger phrases such as “AI 意识”, “保密意识”, “合规意识”, and “不明白就问”, which are common in ordinary discussion and can cause the skill to activate outside clearly intended contexts. Because this is a conversation-level behavioral framework that claims to influence every reply, action, and decision, accidental invocation could unexpectedly alter assistant behavior, create prompt-surface interference, or overshadow more specific user intent.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.