Back to skill

Security audit

China Trip Guide

Security checks for vulnerabilities and agentic risk

Overview

This is a transparent China travel helper with some overly broad triggers, but I found no hidden code, persistence, credential access, or destructive behavior.

Before installing, consider that this skill may answer generic help, emergency, translation, or taxi requests with China-specific travel information. Verify current visa, payment, hospital, and emergency details from official or local sources, especially because the static guide dates from 2024.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
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 (6)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The metadata explicitly sets the skill language to "en", which imposes a language preference without indicating that users may choose another language. This may violate language or locale policy where user language choice should be offered unless the restriction is clearly justified.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The entry-point trigger "help" is extremely generic and can cause the skill to activate during unrelated conversations, increasing the chance that it intercepts user requests outside its intended travel scope. In a travel assistant that includes emergency and payment guidance, accidental activation could produce irrelevant or misleading advice at the wrong time.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The standalone trigger "emergency" lacks domain scoping and may activate for any emergency-related discussion, not just China travel situations. Because this skill provides emergency numbers and medical/police guidance, unintended activation could override or distract from more appropriate local emergency assistance.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The trigger "translation" is too broad and can match many ordinary requests unrelated to this specific skill. Since the skill's translation module is optimized for travel artifacts like signs, menus, and medicine labels, broad activation may lead to context-inappropriate outputs or unnecessary tool use.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The Emergency Help module uses vague standalone triggers such as "help," "emergency," and "lost," which can collide with many ordinary user utterances. In an emergency-oriented module, accidental invocation is more dangerous because it may surface location-specific safety instructions that are inapplicable or delay better assistance.

Content

No source excerpt is available for this finding.

Vague Triggers

Low
Category
Not specified by scanner
Confidence
82% confidence
Finding

The trigger list for Getting Around includes the single word "taxi", which is broad and could be mentioned in many contexts not intended to invoke this module. The file does not define boundaries or exclusion conditions for this trigger.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.