ClawHub
Back to skill

Security audit

GenAI Security & Optimization Gateway

Security checks across malware telemetry and agentic risk

Overview

This is a prompt-security auditing skill, and the risky-looking phrases are disclosed test inputs rather than hidden instructions.

Before installing, note that the marketplace artifact is documentation-only and references runtime files that are not packaged here. Review the source repository before running pip installs or MCP services, and expect a local model download/cache unless you configure a local model and local-only mode.

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

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.prompt_injection_instructions

Prompt-injection style instruction pattern detected.

Warn
Code
suspicious.prompt_injection_instructions
Location
SKILL.md:13