ClawHub
Back to skill

Security audit

Council

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed multi-perspective feedback skill whose main caveat is that user-provided content is sent to a spawned model sub-agent.

This appears safe to install for its stated purpose. Use it only with ideas or documents you are comfortable sending through your model session, invoke it explicitly, and review any custom .md files added to the agents folder because those files become future council-member instructions.

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
83% confidence
Finding
The documented trigger phrases are broad, natural-language commands that could plausibly appear in ordinary conversation, increasing the chance of accidental activation. In a skill that spawns sub-agents and performs follow-up analysis, unintended invocation can consume resources, leak contextual user content into additional agent contexts, or create confusing autonomous behavior.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The example trigger phrases are very broad and map closely to ordinary user requests for advice or feedback. That increases the chance of accidental invocation, causing the skill to spawn sub-agents and process content in a mode the user did not explicitly intend, which can lead to unnecessary data exposure to additional personas/tools and unexpected cost or latency.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The implementation notes define broad trigger phrases without sufficient constraints or counterexamples, making unintended activation likely in normal conversation. In this skill, accidental invocation is more concerning because activation causes multi-agent fan-out and auto-discovery behavior, which amplifies resource usage and may propagate sensitive user content to multiple sub-agents without clear consent.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.