ClawHub
Back to skill

Security audit

unreal-plugin

Security checks across malware telemetry and agentic risk

Overview

The skill is a disclosed Unreal Editor control bridge, but it exposes broad project-changing editor authority without runtime confirmation controls.

Install only for trusted local Unreal projects, keep the project under source control, and require explicit user confirmation before actor deletion, level save/open, console execution, asset import, input simulation, screenshots, or log sharing. Treat it as a powerful editor-control bridge, not a read-only helper.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
86% confidence
Finding
The README encourages broad natural-language use such as creating actors, starting play mode, taking screenshots, and moving scene objects, but does not state hard trigger constraints, confirmation requirements, or clear safety boundaries in the usage section. In the context of a skill that can perform destructive editor actions and run console commands, this increases the risk of unintended or overly permissive activation leading to project modification, disruptive editor actions, or exposure of project content.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The plugin exposes a high-risk command execution bridge into a live Unreal Editor session, including destructive capabilities like actor deletion, level save/open, console execution, asset import, and input simulation, but the tool registration and execution path do not enforce any confirmation or user-visible warning before dispatching those actions. In an agent setting, this increases the chance that an LLM or connected workflow performs irreversible editor actions without the user's informed consent, especially since the code simply queues any requested tool to the first available session.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The `unreal_execute` tool exposes a broad set of powerful editor operations, including destructive actions like actor deletion, asset import, level save/open, console execution, and input simulation, but the implementation performs no policy checks or confirmation gating before queueing the command. In this skill context, the plugin is specifically designed to control a local Unreal Editor, so lack of enforcement means an agent or prompt-influenced workflow could trigger irreversible project changes or unsafe editor commands immediately.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.