ClawHub
Back to skill

Security audit

Accountable AI

Security checks across malware telemetry and agentic risk

Overview

This is a mostly documentation-only governance skill, but it gives agents broad authority around deletion, credentials, proactive contact, and repo pushes without enough user-control guardrails.

Review and customize these documents before installation. Remove the automatic bootstrap deletion, replace credential documentation with references to a secrets manager, and require explicit approval before external messages, repo pushes, deployments, or changes to shared configuration.

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
94% confidence
Finding
The document explicitly instructs the agent to delete a bootstrap file after reading it, which is a destructive action performed without confirmation or a clear safety check. In an agent workspace, bootstrap files may contain audit-relevant initialization details, and automatic deletion can remove evidence, break recovery, or be abused to hide malicious setup steps.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.