ClawHub
Back to skill

Security audit

Devil's Advocate (魔鬼代言人)

Security checks across malware telemetry and agentic risk

Overview

This is a prompt-only critique skill that may be blunt and over-triggered, but it does not request code execution, credentials, files, network access, or persistence.

Install this only if you want deliberately blunt critical feedback. Because it may trigger broadly, ask explicitly for a balanced or non-adversarial response when needed, and avoid using it for sensitive decisions unless you want that confrontational review style.

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 (1)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The activation description is broad enough to trigger on many generic planning, feedback, or decision-making conversations, which can cause unintended invocation of a highly opinionated skill. In context, this is risky because the skill enforces a rigid adversarial response style and could override normal assistant behavior in situations where the user did not explicitly request confrontational critique.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.