ClawHub
Back to skill

Security audit

Meituan

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Meituan travel-query connector that uses OOMOL's CLI and does not currently define any state-changing Meituan actions.

Before installing, confirm you trust OOMOL and are willing to connect a Meituan account through OOMOL. Treat the remote CLI installer and persistent account connection as the main trust decisions; current skill behavior is limited to read-style travel queries, and any future tagged write/destructive action should require explicit confirmation.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Description-Behavior Mismatch

Medium
Confidence
92% confidence
Finding
The manifest markets the skill as only for searching and reading data, but the body explicitly defines handling for [write] and [destructive] actions. This mismatch can mislead callers, reviewers, or automated policy layers into treating the skill as read-only when it is designed to support state-changing operations, increasing the risk of unintended or insufficiently guarded writes.

Vague Triggers

Medium
Confidence
86% confidence
Finding
The trigger text says to use this skill for ANY Meituan request, which is overly broad and can cause the agent to invoke it for unrelated, sensitive, or higher-risk Meituan tasks without sufficient scoping. In this skill, that is more dangerous because the documentation contemplates state-changing actions, so an overbroad routing rule could funnel write-capable requests into the skill by default.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.