ClawHub
Back to skill

Security audit

Nm Imbue Review Core

Security checks across malware telemetry and agentic risk

Overview

The skill appears to be a review-workflow guidance skill with broad activation language but no evidence of hidden execution, persistence, data theft, or destructive behavior.

Install only if you want a skill that may be invoked for general review or analysis tasks. If you prefer tighter routing, narrow the trigger phrases before use; otherwise the observed risk is mainly that it could influence more workflows than intended.

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

High
Confidence
95% confidence
Finding
The skill defines very broad trigger keywords and an auto-load condition that can activate on almost any review or analysis request. This creates an excessive invocation surface, increasing the chance the skill is loaded in unintended contexts and influences workflows where it was not specifically requested. In an agent environment, overbroad auto-activation can cause prompt-routing confusion, scope creep, and accidental precedence over more appropriate specialized skills.

Vague Triggers

High
Confidence
93% confidence
Finding
The manifest trigger list uses generic terms like workflow, evidence, reporting, and analysis without qualifiers, making the skill eligible for activation across many unrelated tasks. Because this skill provides process scaffolding rather than a narrowly bounded function, generic triggers can unintentionally alter agent behavior broadly, especially when multiple skills compete for selection. The context makes this somewhat more dangerous because the skill is designed to run at the start of reviews, giving it early influence over task framing and evidence collection.

VirusTotal

47/47 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.