ClawHub
Back to skill

Security audit

AI视频剪辑工具箱｜AI-HIVE

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent AI-HIVE video editing helper that uses local video tools and user-directed API calls, with no artifact-backed deception or unrelated data access found.

Install only if you are comfortable using AI-HIVE for media generation. Review commands before running them, upload only media you have rights to use, confirm cost-related generation settings first, and keep API keys out of logs, screenshots, and repositories.

SkillSpector

Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Lp3

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding
The skill includes executable code paths and operational guidance that use environment variables, local file access, shell commands, and network calls, but it does not declare permissions or capability boundaries. This creates a trust and review gap: a host or user may invoke the skill without understanding that it can access local media files, call external services, and rely on secrets such as API keys.

Static analysis

No suspicious patterns detected.