ClawHub

Zenith Man

Security checks across malware telemetry and agentic risk

Overview

This appears to be a disclosed instruction-only book/legal-help skill with overbroad activation wording, but no evidence of malware, credential access, persistence, or destructive behavior.

Reasonable to install if you want this book/framework to guide relevant legal-help conversations. Be aware it may activate too broadly, so users should explicitly ask the agent to stop using the framework when it feels off-topic and should not treat it as a substitute for qualified legal advice.

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
94% confidence
Finding
The trigger list is broad, loosely phrased, and includes common legal/help-seeking language plus generic onboarding conditions like users saying they just installed the skill or do not know how to start. This can cause unintended activation in unrelated conversations, leading the agent to inject legal-defense framing or book-specific guidance when the user did not explicitly request it.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The Quick Start instructs the AI to appear whenever it 'senses this book could help,' which delegates invocation to an undefined, highly subjective heuristic. In practice, this increases prompt-surface area and raises the chance of unsolicited or miscontextualized responses, especially in sensitive legal discussions where the model may over-apply the skill's worldview.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal