ClawHub
Back to skill

Security audit

Agent Hardening

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent agent-hardening test pack that uses visible synthetic examples and does not request sensitive access.

Reasonable to install if you want agent hardening test examples. Treat the embedded adversarial strings as test data, review the visible Python snippets before running them, and re-check future versions for any added file, network, credential, or persistence behavior.

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:44