ClawHub

ArifOS Sense

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed governance-review skill that may interrupt normal use and keep local audit logs, but it shows no hidden code, exfiltration, or destructive behavior.

Install only if you want a persistent governance layer that can pause high-stakes actions and record local audit entries. Review the ledger path and trigger list first, especially if broad HOLD/VOID behavior or local retention of decision context would be disruptive.

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 frontmatter trigger list includes highly generic terms such as "evaluate," "hold," and "is this safe," which are likely to appear in many unrelated conversations. In a governance skill that can alter agent behavior and block or redirect actions, overly broad activation creates a real risk of unintended interception, policy confusion, or denial-of-service-like overactivation.

Vague Triggers

Medium
Confidence
96% confidence
Finding
The markdown trigger section repeats broad activation criteria like common words and vague situations, making it ambiguous when the skill should engage. Because this skill is framed as a "non-negotiable governance kernel" with authority to pause or veto actions, ambiguous triggers materially increase the chance that benign requests are escalated into HOLD/VOID flows or that user intent is misclassified.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal