ClawHub
Back to skill

Security audit

travel-guide（本地土著旅游攻略）

Security checks across malware telemetry and agentic risk

Overview

This is a plain travel-planning helper with broad routing keywords, but it does not run code, request credentials, or persist data.

Install it if you want a travel-planning reference skill, but expect it may activate for broad travel queries and may ask about budget, dates, and party details. Avoid sharing sensitive identifiers such as passport or ID numbers, and verify time-sensitive facts like visa rules, safety advisories, weather, and prices before booking.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (4)

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger list is extremely broad and includes many generic phrases like travel recommendations, planning, hotels, weather, and where to go. This can cause the skill to activate for ordinary conversation more often than intended, increasing the chance of misrouting users into this skill and overriding more appropriate domain-specific handling.

Natural-Language Policy Violations

Medium
Confidence
79% confidence
Finding
The skill content strongly implies bilingual Chinese-English behavior throughout the header and body without giving the user a language preference path. This can degrade usability, produce unnecessarily verbose responses, and create confusion or disclosure risk if a user expects a single-language interaction.

Vague Triggers

Low
Confidence
86% confidence
Finding
The manifest includes a very broad set of generic travel-related keywords, which can cause the skill to activate for many common user requests without strong scoping. In a travel skill this is not inherently malicious, but it increases the chance of unintended routing, over-collection of user travel context, or the skill responding where a narrower skill would be more appropriate.

Natural-Language Policy Violations

Low
Confidence
69% confidence
Finding
The description presents bilingual Chinese/English behavior as a default without stating that output language should follow user preference. This can lead to unnecessary disclosure or confusing responses in mixed-language contexts, and may degrade usability or cause accidental inclusion of content the user did not request.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.