ClawHub
Back to skill

Security audit

智能景点推荐

Security checks across malware telemetry and agentic risk

Overview

This travel recommendation skill uses external travel and map proxy APIs in a disclosed, purpose-aligned way, with privacy considerations around itinerary and location queries.

Install only if you are comfortable sending travel searches, route origins/destinations, dates, and location keywords through the skill publisher's cloud proxy to Fliggy/Gaode-backed services. Avoid entering private addresses or sensitive itinerary details unless that data sharing is acceptable to you.

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
85% confidence
Finding
The skill declares no permissions, yet its own metadata describes outbound network use via a cloud proxy and upstream Fliggy/Amap APIs. That mismatch reduces transparency and prevents proper consent/review of external data flows, which can expose user itinerary queries and related travel data to third-party services or an intermediate proxy.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill sends user-supplied travel data, including queries, origins/destinations, and location-related inputs, to two hardcoded third-party proxy endpoints via HTTP requests without any visible consent, disclosure, or data-minimization controls. In a travel assistant context this can expose sensitive itinerary and location data to external operators, and the risk is increased because the code also embeds a static proxy token and relies on opaque proxy infrastructure rather than direct, documented first-party APIs.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.