ClawHub
Back to skill

Security audit

AI短剧制作助手 | AI Short Film Producer

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only AI video production skill with expected third-party API, credential, cost, and local file-output considerations.

Before installing, be comfortable sending prompts, scripts, voice text, and possibly media to third-party services. Use a scoped or low-balance API key, confirm costs and retry limits before batch jobs, choose the export folder deliberately, and use temporary or access-limited public links for any media that must be reachable by an API.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger examples are broad enough to match routine requests like general short-video creation or turning text into video, which can cause the skill to activate outside the user's likely expectations. In this skill, unintended activation is more concerning because the documented workflow includes external API calls, local media generation, and desktop file exports, increasing the chance of unanticipated data sharing or filesystem side effects.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill describes sending user-provided content to third-party video and TTS APIs and writing outputs to local desktop folders, but it does not include user-facing notice, consent, or safety boundaries for those actions. In context, this is materially risky because scripts, voice text, and reference materials may contain sensitive data, and local export behavior can surprise users or expose private project contents on shared machines.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.