ClawHub
Back to skill

Security audit

LLM Council

Security checks across malware telemetry and agentic risk

Overview

This skill transparently runs a multi-agent deliberation workflow and does not include hidden code, persistence, credential use, or unrelated data access.

Before installing, be aware that this skill can spend extra time and model calls by dispatching multiple advisors and reviewers. It is best used for genuinely high-stakes or ambiguous decisions, not routine questions.

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
89% confidence
Finding
The trigger list includes very broad natural-language phrases such as "should I", "am I right about", and "council", which are likely to appear in many ordinary user prompts that do not actually request this specific multi-agent deliberation workflow. That can cause unintended invocation, increasing cost, latency, and the chance that a more invasive or overcomplicated process is applied where a simple answer was appropriate.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.