ClawHub
Back to skill

Security audit

Shopee-类目蓝海挖掘专家 | 青虎AI

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent Shopee category-research helper that uses a disclosed Qinghu API token and requires user approval before paid data calls.

Install this if you want Qinghu-powered Shopee category research. Be prepared to provide or expose a Qinghu API token, confirm external API calls before they run, and monitor Qinghu point usage; for non-Shopee marketplace strategy, choose a more specific skill instead.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

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
94% confidence
Finding
The metadata description says the skill 'must trigger' for broad category-selection requests such as opening a new store, evaluating market capacity, or finding high-growth/low-competition categories. Those phrases are common in general ecommerce conversations, so the skill may activate outside a narrowly scoped Shopee-analysis context and overtake more appropriate tools or workflows. The mandatory wording increases the risk of misrouting user requests and causing unnecessary external API use or charges.

Vague Triggers

Low
Confidence
89% confidence
Finding
The '何时触发' examples enumerate several user utterances that should activate the skill, but they do not define exclusion criteria or non-trigger cases. In a skill that can request credentials, call paid external tools, and auto-export data, missing boundaries increases the chance of over-activation on ambiguous requests and unnecessary data/tool usage. The context makes this more concerning than a harmless FAQ skill because misfires can drive paid actions and collect tokens.

Static analysis

No suspicious patterns detected.