ClawHub
Back to skill

Security audit

抖音爆款封面生成与编辑

Security checks for vulnerabilities and agentic risk

Overview

The skill performs a disclosed AI Hive image-generation workflow and I did not find hidden, destructive, or deceptive behavior.

Install only if you intend to use AI Hive for Douyin/ecommerce cover generation. Treat the API key as sensitive, avoid uploading confidential images or proprietary prompts, and review the output directory before using generated assets publicly.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
94% confidence
Finding
The skill instructs use of a local Python script with network, shell, filesystem, and environment access, but it declares no permissions or user-facing capability boundaries. This creates a transparency and consent problem: users may invoke a seemingly narrow image-cover skill without understanding that it can access credentials, read local files passed as inputs, write outputs, and make outbound API calls.

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding
The documented purpose is limited to Douyin cover generation, but the underlying referenced tooling appears capable of much broader actions including chat, video generation, generic media upload, account-info retrieval, wallet/balance access, model enumeration, and browser-based API-key setup. That mismatch is dangerous because it obscures the real attack surface and can enable over-privileged or unexpected operations under the guise of a simple design skill.

Vague Triggers

Medium
Confidence
81% confidence
Finding
The activation language is broad enough to match many generic image, thumbnail, poster, first-frame, and e-commerce visual requests, which can cause the skill to trigger outside the user's specific intent. Over-broad triggering becomes a security concern when the skill can then invoke scripts, upload files, or initiate outbound API activity without the user clearly requesting this particular workflow.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill documents use of API keys via environment variable or local config and references networked task execution, but it does not clearly warn users about credential sensitivity, storage location risks, or outbound transmission to a third-party service. This increases the chance of accidental secret exposure, unsafe local storage, or users unknowingly sending proprietary images and prompts to an external API.

Static analysis

No suspicious patterns detected.