ClawHub
Back to skill

Security audit

长视频转30条短视频｜AI-HIVE

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed video-repurposing helper that uses AI-HIVE and local ffmpeg tools, with no evidence of hidden exfiltration, destructive behavior, or automatic background activity.

Install only if you are comfortable using AI-HIVE for media generation and uploading authorized media to that service. Treat generated calls as potentially billable, review parameters before running them, and avoid storing or pasting real API keys into logs, screenshots, 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
92% confidence
Finding
The skill exposes operational guidance for environment-variable use, local file I/O, shell execution, ffmpeg processing, and outbound network access, yet declares no permissions or equivalent capability boundaries. This creates a transparency and trust gap: an agent or reviewer may approve the skill under a lower-risk assumption while it can still trigger file, shell, and network behaviors that could access sensitive local data or perform unintended external actions.

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding
The documented purpose is a specialized long-video-to-30-shorts workflow, but the described/code-linked behavior appears to include much broader generic AI-HIVE capabilities: arbitrary model querying, text/image/video generation, media upload, account/balance access, and generic ffmpeg/media operations. This mismatch is dangerous because it can mislead users and policy systems about the true authority of the skill, enabling capability creep, unexpected external API usage, privacy exposure, or billable actions outside the advertised narrow use case.

Vague Triggers

Medium
Confidence
93% confidence
Finding
Enabling implicit invocation without trigger constraints can cause the skill to activate in broader contexts than intended, including ambiguous requests that only partially match the skill’s purpose. Because this skill can generate production-ready workflows and runnable AI-HIVE commands for marketing/video repurposing, over-broad auto-invocation increases the chance of unintended execution paths, misuse for disallowed content-generation scenarios, or accidental application to unauthorized media tasks.

Static analysis

No suspicious patterns detected.