ClawHub
Back to skill

Security audit

Phy Ui Reviewer

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only UI code review checklist with no executable code, credential requests, persistence, or hidden data handling.

This skill appears safe to install for UI review tasks. Before using it on private code, remember that the agent will need access to the files being reviewed, and expect reports to follow the Chinese-language template unless the user or host overrides language preference.

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)

Natural-Language Policy Violations

Medium
Confidence
96% confidence
Finding
The skill metadata and body are written to force Chinese-language behavior without indicating that the language should follow user preference or runtime locale. This can override user intent, reduce usability, and create prompt-steering behavior where the agent responds in an unexpected language, which is a real quality and policy issue even though it is not a code-execution risk.

VirusTotal

No VirusTotal findings

View on VirusTotal

Static analysis

No suspicious patterns detected.