ClawHub
Back to skill

Security audit

邮轮搜索

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed cruise-search integration that sends travel search parameters to a Tuniu proxy and shows no hidden local data access, persistence, or destructive behavior.

Before installing, understand that cruise searches, departure cities, dates, route keywords, and product IDs are sent to the skill’s Tencent Cloud proxy and then to Tuniu. Avoid entering sensitive personal information in search text. The publisher should handle the proxy token through secure configuration rather than embedding a fallback token.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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)

Lp3

Medium
Category
MCP Least Privilege
Confidence
87% confidence
Finding
The skill declares and relies on environment-backed network access to a proxy API, but the metadata shown does not declare explicit permissions for those capabilities. This can weaken platform trust boundaries and user/operator awareness, because the skill can transmit user queries and access secrets without a clear permission contract. In this context the risk is moderated by the skill’s stated travel-search purpose, but it still exposes external data flow and secret use.

Env Variable Harvesting

High
Category
Data Exfiltration
Content
]

PROXY_URL = os.environ.get("PROXY_URL", "https://1439498936-0junm3maxj.ap-guangzhou.tencentscf.com")
PROXY_TOKEN = os.environ.get("PROXY_TOKEN", "tp_8k2mX9vQ4z")


def _validate_proxy_url(url):
Confidence
98% confidence
Finding
os.environ.get("PROXY_TOKEN

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.