ClawHub
Back to skill

Security audit

OGT Docs Define

Security checks across malware telemetry and agentic risk

Overview

This is a documentation workflow skill whose local file changes are limited to expected status marker files, with no hidden network, credential, or persistence behavior found.

Install this if you want help organizing definition documents. Before running the example shell commands, confirm you are in the intended definition directory because the workflow creates and removes local marker files.

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
92% confidence
Finding
The skill includes shell snippets with destructive file operations such as `rm .draft` and `rm .ready_for_review .review_requested_at` without any cautionary note, validation step, or safer alternative. In an agent-skill context, users or downstream agents may copy or execute these commands mechanically, and although the scope appears intended for local signal files, the pattern normalizes deletion commands and can cause accidental data loss if run in the wrong directory or adapted unsafely.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.