ClawHub
Back to skill

Security audit

电商爆款带货视频生成与编辑

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed AI Hive/Seedance e-commerce video workflow that uses user-provided assets and an API key for generation, with no hidden persistence or unrelated behavior found.

Before installing, be aware that using the generation commands will upload the product images or videos you provide to AI Hive and may use or store an AI Hive API key. Use --preview first when you want to inspect the prompt without uploading or being charged, and only provide assets and credentials you are comfortable sending to that service.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The skill instructs users to run a local Python script that can read files, write outputs, access environment variables and API keys, make network requests, and invoke shell commands, yet no permissions are declared. This creates a transparency and trust gap: users may invoke the skill without understanding that it can access local paths, stored credentials, and external services, increasing the risk of unintended data exposure or unsafe execution.

Vague Triggers

Medium
Confidence
82% confidence
Finding
The description is intentionally broad and includes many platforms, use cases, and competitor/tool names, which can cause the skill to trigger on generic user requests about video generation or alternative tools. Over-broad invocation increases the chance of inappropriate routing, where users are funneled into this skill and encouraged to execute its code path even when they asked for general advice rather than local script execution.

Natural-Language Policy Violations

Medium
Confidence
88% confidence
Finding
The skill mandates a fixed model and workflow ('固定使用 Seedance 2.5') without user opt-in, which can override user intent, cost preferences, privacy expectations, or provider choice. In this context, forced provider selection matters because the workflow involves uploading local product assets and potentially using stored API keys, so users may be routed to an external service they did not explicitly choose.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.