ClawHub
Back to skill

Security audit

AI电商上架套图｜AI-HIVE

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed AI-HIVE helper for planning and generating e-commerce listing images, with user-controlled API use and no evidence of hidden exfiltration or destructive behavior.

Installers should understand that this skill can call AI-HIVE with an API key, upload files they explicitly provide as references, and download generated outputs. Confirm prompts, routing, batch size, costs, and media authorization before running generation, and avoid storing or sharing API keys in logs or repositories.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding
The skill advertises executable workflows that use environment variables, local file access, network calls, and shell commands, yet no permissions are declared. That creates a transparency and containment problem: a caller or host may invoke a skill believing it is low-risk documentation, while it can actually access local resources and external services.

Tp4

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding
The stated purpose is narrowly about e-commerce listing image-set generation, but the detected behavior expands into generic chat, video generation, wallet/user-info access, model enumeration, uploads, and local credential storage. This mismatch undermines user consent and review boundaries, because a seemingly narrow skill can expose credentials, query account data, or perform broader networked actions than expected.

Vague Triggers

Medium
Confidence
78% confidence
Finding
The activation criteria extend beyond concrete trigger phrases into broad requests like turning reference cases into original content and stable delivery for advertising/social scenarios. Overbroad activation increases the chance the skill runs in contexts the user did not intend, which is more concerning here because the skill also interfaces with external APIs and potentially billable generation workflows.

Static analysis

No suspicious patterns detected.