ClawHub
Back to skill

Security audit

takeout-decision-fatigue-rescuer

Security checks for vulnerabilities and agentic risk

Overview

This skill is a low-risk takeout recommendation helper that asks for meal preferences and does not include code, persistence, credentials, or hidden actions.

Before installing, be aware that the skill may activate on casual meal-indecision phrases. It may ask for budget, dietary restrictions, hunger level, and possibly location or local restaurant context if real takeout recommendations are desired; no sensitive permissions or automatic actions are evident.

SkillSpector

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 (2)

Vague Triggers

Medium
Confidence
87% confidence
Finding
The skill’s trigger phrases are very broad everyday expressions like '不知道吃什么' and '中午吃啥', which can cause the skill to activate in contexts where the user is casually chatting rather than explicitly requesting structured recommendation help. This is dangerous because unintended invocation can derail the conversation, collect unnecessary preference data, or override a more appropriate skill, even though the domain itself is low risk.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger section defines when to activate but does not specify when not to activate, leaving the skill without exclusion boundaries. In practice, this can cause the skill to capture adjacent intents like general meal planning, restaurant discussion, or emotional venting about food apps, creating reliability and consent issues through overbroad routing.

Static analysis

No suspicious patterns detected.