ClawHub
Back to skill

Security audit

AI大模型专家｜AI漫剧生成与编辑

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed AI-HIVE media-generation helper that uses a user-provided API key and selected media files; I found no hidden exfiltration, destructive behavior, or automatic privileged execution.

Install only if you intend to use AI-HIVE for media generation and are comfortable storing an AI-HIVE API key locally. Treat any referenced images, videos, or audio as uploads to the provider, verify you have rights to the materials, and review generated commands before running them because generation jobs may consume account credits.

SkillSpector

By NVIDIA
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 (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding
The skill advertises and demonstrates shell execution, environment variable use, local file reads/writes, and network access, but does not declare permissions or clearly scope those capabilities. This creates a transparency and containment problem: a host or user may authorize the skill based on incomplete metadata, while the included workflows can access local secrets, read/write files, and send data to external services.

Tp4

High
Category
MCP Tool Poisoning
Confidence
84% confidence
Finding
The documented behavior extends beyond the narrowly described AI comic/short-drama workflow into generic chat, model inspection, account/balance querying, media upload, API-key setup, and unrelated image tooling. This mismatch is dangerous because users may trust the skill with credentials and content under a specialized-use assumption, while the actual capability surface is broader and can expose sensitive data or perform unexpected external actions.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The default prompt uses a very broad invocation phrase tied to a general project request, without clear scope boundaries or trigger constraints. This increases the chance the skill is activated in contexts the user did not explicitly intend, which can cause over-broad delegation, unintended data exposure to the skill workflow, or prompt steering into external tool usage.

Vague Triggers

Medium
Confidence
95% confidence
Finding
Enabling implicit invocation without narrowing context allows the system to auto-select this skill based on loosely matching user language. In a skill that can orchestrate content generation workflows and handle uploaded media via external model infrastructure, this raises the risk of unintended activation, unnecessary sharing of sensitive project materials, and user confusion about which agent is acting.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical
Code
suspicious.exposed_secret_literal
Location
SKILL.md:127