Back to skill

Security audit

邮轮搜索工具

Security checks for vulnerabilities and agentic risk

Overview

This is a remote cruise-search skill with incomplete documentation, but no evidence of hidden local access, persistence, destructive behavior, or malware.

Install only if you are comfortable sending cruise-search queries to the listed CruiseSkillBridge/olavacations remote service. Avoid entering sensitive personal, booking, account, or payment details unless the publisher documents how that data is handled.

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Intent-Code Divergence

Medium
Confidence
90% confidence
Finding
The documentation says the tool is for cruise product search, but the usage section instructs users to submit a generic 'analysis request' to an external endpoint. This mismatch is risky because it obscures the tool’s true behavior, can mislead operators about what data is being sent, and may indicate copied or placeholder integration instructions that were never validated for this skill’s stated purpose.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill instructs users to send input to external APIs, including a gateway URL, but provides no notice that user data will leave the local environment or be processed by third parties. This is dangerous because users or downstream agents may transmit sensitive search terms, identifiers, or other confidential content without informed consent or any documented data-handling constraints.

Static analysis

No suspicious patterns detected.