Security audit

Wa Quanquaner Skill

Security checks for vulnerabilities and agentic risk

Overview

This coupon skill mostly does what it says, but it can activate on generic food/chat phrases and asks for broader command permissions than a coupon lookup needs.

Review before installing if you do not want a skill that may surface coupon links during broad food-delivery or meal-choice conversations. Prefer narrowing activation to explicit coupon or discount requests and removing unrestricted PowerShell permission unless there is a documented need.

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 (3)

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The skill advertises very broad natural-language triggers such as '今天吃什么', '点外卖', and other common everyday requests, which can cause the agent to invoke this skill in many ordinary conversations unrelated to coupon retrieval. Over-broad activation increases the chance of unintended routing, user confusion, and opportunistic promotion of external links, especially because the skill centers on redirecting users to third-party offer pages.

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The description includes broad everyday phrases like '今天吃什么', '点外卖', and '叫外卖', which can cause the skill to activate in many ordinary conversations unrelated to coupon retrieval. Over-broad invocation increases the chance of unsolicited tool execution and unnecessary outbound requests to the external coupon service.

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The dedicated trigger section repeats ambiguous activation keywords without constraints, again including generic phrases that are common in normal meal-planning chat. In context, this is more risky because the skill is configured to use network-capable tools, so accidental invocation can produce external requests and promotional output the user did not explicitly ask for.

Static analysis

No suspicious patterns detected.