ClawHub

旅行美食助手

Security checks across malware telemetry and agentic risk

Overview

This travel food skill is coherent and disclosed, but live restaurant searches send place queries to an external map proxy.

Install only if you are comfortable with searched place names, addresses, cuisine keywords, and nearby-search parameters being sent to the skill publisher’s proxy-backed map service. Avoid using it in an environment where PROXY_TOKEN contains an unrelated secret.

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
89% confidence
Finding
The skill instructs the agent to execute a Python script and explicitly states that query parameters are sent to an external proxy service and then to a map POI service, which confirms network-capable behavior. Because no permissions are declared despite using code capabilities such as network and possible environment access, users and reviewers are not given transparent notice of the skill’s data access and outbound communication, increasing the risk of unintended data exposure or abuse.

Missing User Warnings

Medium
Confidence
81% confidence
Finding
User-supplied location queries are transmitted to an external proxy/geolocation service without any disclosure or consent mechanism. In a travel assistant, location data is especially sensitive because it can reveal itinerary, current whereabouts, and habits, and users may not realize their inputs leave the local skill boundary.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal