ClawHub
Back to skill

Security audit

文章转视频 Article to Video

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed dLazy SaaS wrapper for turning articles or documents into videos, with expected file upload and API-key use.

Install only if you are comfortable sending prompts and any attached documents to dLazy's hosted service and storing a dLazy API key locally. Avoid uploading sensitive documents unless your organization approves dLazy for that data.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Description-Behavior Mismatch

Medium
Confidence
89% confidence
Finding
The skill is presented as narrowly scoped to article-to-video conversion, but the runtime is explicitly pinned to a broader `file-to-video` template. This scope mismatch can cause an invoking agent or user to send a wider class of local documents than expected, increasing the chance of unintended data exposure to the hosted SaaS backend.

Intent-Code Divergence

Medium
Confidence
86% confidence
Finding
The documentation repeatedly brands the skill as `article-to-video`, while the operational instructions direct the agent to invoke `dlazy chat --skill file-to-video`. That inconsistency is security-relevant because agents rely on skill labels for trust and data-sharing decisions; misleading scope can result in broader document processing and upload than the user reasonably intended.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.