ClawHub
Back to skill

Security audit

Cross-Border-Ecommerce-Product-Picker

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed ecommerce product-research helper that uses web browsing/search for public trend and pricing research, with no code execution, persistence, or payment authority.

Before installing, understand that the skill may perform live web searches when ecommerce category or product-selection language is used, and its estimates should be treated as research aids rather than guaranteed sales or profit numbers. Non-Chinese users may need to ask the agent to respond in another language.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

High
Category
Not specified by scanner
Confidence
96% confidence
Finding

The trigger condition is very broad and uses generic phrases like “帮我选品”, “分析类目”, and category mentions such as “选宠物用品”, which can cause the skill to auto-activate in contexts where the user did not explicitly request this specific tool. In a skill with browser and web_search access, unintended invocation increases the chance of unnecessary external lookups, irrelevant guidance, and context hijacking from loosely related conversations.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The instruction to always respond in Chinese overrides user preference and can reduce transparency, comprehension, and informed consent for users expecting another language. While not a direct code-execution or data-exfiltration issue, forced language constraints can make the skill behave unexpectedly and increase the risk of user misunderstanding around estimates, links, and commercial recommendations.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.