Generator Editing Generator

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:40 PM UTC
• Version:
  1.0.0

Overview

This is a cloud video-editing skill whose network use and media upload behavior match its stated purpose, though users should understand their videos and prompts go to NemoVideo’s backend.

Install only if you are comfortable sending uploaded videos, edit prompts, timeline data, and render requests to mega-api-prod.nemovideo.ai. Use a dedicated NEMO_TOKEN if available, avoid sensitive footage unless you trust the provider’s handling practices, and be cautious with generic video requests because this skill’s trigger wording is broad.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (4)

Context-Inappropriate Capability

Medium

• Confidence: 89% confidence
• Finding: The skill instructs the agent to automatically mint and use an anonymous token that grants free credits and establishes an authenticated backend session. That expands the skill from simple media processing into account/session acquisition and consumption of a third-party service, which can be abused for unauthorized resource use, opaque account creation, and hidden billing or quota impacts if done without explicit user consent.

Vague Triggers

Medium

• Confidence: 80% confidence
• Finding: The phrase "Or just tell me what you're thinking" is broad enough to match ordinary conversation and may cause the skill to activate when the user did not intend to invoke a cloud video-editing workflow. In this skill, accidental invocation is more concerning because it can lead to remote processing setup and potential upload/session actions against an external service.

Vague Triggers

Medium

• Confidence: 91% confidence
• Finding: The catch-all routing rule sends "Everything else" to the SSE editing path, creating an overly permissive trigger surface. Because that path can transmit user text to a remote backend and potentially alter session state, ambiguous prompts may be misrouted into external processing without clear user intent.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The skill states that user videos are sent to a cloud processing backend and remote GPU nodes, but it does not require a clear, affirmative warning at the moment of upload or processing. Users may unknowingly transmit potentially sensitive media off-device, creating privacy, confidentiality, and compliance risks.

VirusTotal

65/65 vendors flagged this skill as clean.