ClawHub
Back to skill

Security audit

Pentest Commands

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed, instruction-only penetration-testing command reference, but its commands should only be used on systems where the user has explicit authorization.

Install this only if you need a penetration-testing command reference. The skill does not run tools by itself, but the commands it provides can scan, exploit, brute-force, generate payloads, and dump data; use them only within written authorization and a clearly defined assessment scope.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Confidence
86% confidence
Finding
The skill is a broad offensive command reference that includes exploitation, brute forcing, payload generation, reverse shells, and lateral-movement-related actions, but its safety language is limited to brief constraints rather than strong user-facing guardrails. In an agent setting, this can enable harmful dual-use activity by making dangerous actions easy to invoke without meaningful friction, context checks, or authorization enforcement.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.