ClawHub
Back to skill

Security audit

XiaoChi

Security checks across malware telemetry and agentic risk

Overview

This instruction-only skill is limited to summarizing public Meituan Waimai merchant and product information, with no evidence of hidden code or account-changing behavior.

Reasonable to install for public Meituan Waimai research. Avoid entering a precise home address unless needed, confirm time-sensitive prices and delivery estimates, and do not use it for ordering, scraping at scale, bypassing platform controls, or accessing non-public account data.

SkillSpector

By NVIDIA
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 (1)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The invocation description is overly broad: "当需要访问或自动化美团外卖相关内容时调用" can match many ordinary user requests about Meituan Waimai, causing the skill to trigger outside its narrowly described scope. Over-broad routing increases the chance of unnecessary web access or automation on a third-party platform and can bypass more appropriate, lower-risk handling for simple informational queries.

VirusTotal

60/60 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.