ClawHub
Back to skill

Security audit

Skill Firewall

Security checks across malware telemetry and agentic risk

Overview

The skill is a defensive prompt-injection review helper, with no artifact evidence of hidden execution, persistence, or data access beyond reviewing external skill text.

Before installing, understand that this skill may step in broadly when external skills are discussed. That can be useful for security review, but you may want to invoke it only when reviewing, importing, or rewriting an external skill.

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
84% confidence
Finding
The description says to use this skill whenever external skills are mentioned, which is broader than necessary and can cause the skill to activate on casual discussion rather than actual install/use flows. Overbroad triggering increases the chance of unintended interception, prompt steering, or workflow disruption across unrelated conversations.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.