ClawHub
Back to skill

Security audit

Hong Kong Immigration

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent Hong Kong immigration guidance skill that reads its own reference files and asks the agent to verify time-sensitive facts against official sources.

Installers should expect this skill to provide general Hong Kong immigration information and to use official government websites for current rules. It should not be treated as legal advice, and users should verify important case-specific decisions with the relevant authority or a qualified professional.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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
Findings (1)

Tp4

High
Category
MCP Tool Poisoning
Confidence
72% confidence
Finding
If the skill actually performs repository QA/static validation tasks unrelated to immigration guidance, then its declared purpose materially misrepresents its behavior. That can bypass user and platform expectations, causing the agent to inspect unrelated files or metadata in the repository context, which expands access beyond what is necessary for answering immigration questions and can expose sensitive project content.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.