ClawHub
Back to skill

Security audit

GStack Review Stack

Security checks across malware telemetry and agentic risk

Overview

This is a coherent review-helper skill; its only notable risk is a disclosed, user-triggered fallback to inspect a public GitHub repository for upstream reference material.

Install if you want structured review and ship-gate guidance. When asking it to use upstream gstack, expect your agent may contact GitHub and read public repository content; for stricter environments, prefer a local or pinned clone and treat upstream instructions as reference material rather than trusted commands.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Description-Behavior Mismatch

Low
Confidence
89% confidence
Finding
The skill instructs fetching or inspecting an external GitHub repository when local references are missing, but it is presented as a review tool rather than a network-capable skill. That creates a supply-chain and data-governance risk because execution may pull unpinned remote content into the decision process without explicit user consent or trust validation.

Context-Inappropriate Capability

Low
Confidence
86% confidence
Finding
Remote repository fetch/inspection is broader than necessary for a review skill unless tightly scoped, and it can introduce untrusted external instructions or content into the workflow. In this context, the danger is not code execution from SKILL.md itself but unreviewed dependency on mutable external material that can influence outputs or trigger unexpected network behavior.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The instructions allow fetching or inspecting an external GitHub repository without warning the user that network access may occur. Hidden or unexpected network activity is dangerous because it can violate user expectations, organizational policy, or privacy constraints, especially in environments where outbound access is sensitive or logged.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.