Vertical Video Editor

• Outcome: Pass
• Latest audit: May 29, 2026, 8:32 PM
• Version:
  1.0.2

Overview

This is a cloud-based video editing skill that clearly relies on NemoVideo services, with privacy-sensitive but purpose-aligned media upload and session setup behavior.

Install only if you are comfortable using NemoVideo as a cloud processor for your clips. Avoid uploading confidential, regulated, client-owned, or unreleased footage unless NemoVideo's privacy and retention terms are acceptable, and prefer a dedicated or revocable NEMO_TOKEN when using your own credentials.

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
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (3)

Description-Behavior Mismatch

Medium

• Confidence: 93% confidence
• Finding: The skill is presented like a straightforward video-editing utility, but it actually instructs the agent to provision tokens, create remote sessions, check credits, generate claim links, and export through a third-party backend. This materially expands the data flow and trust boundary beyond what a user would reasonably expect from the description, increasing the risk of unintended disclosure of media, metadata, and session identifiers.

Context-Inappropriate Capability

Medium

• Confidence: 87% confidence
• Finding: Allowing uploads by arbitrary URL gives the skill a network-fetch capability beyond simple handling of user-supplied local media. If abused, this can be used to pull in unexpected remote content, including internal or sensitive URLs if the surrounding platform permits it, creating SSRF-like or data exfiltration risk.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill directs automatic connection to a remote backend on first open and implies transmission of user media and metadata without an explicit opt-in or prominent warning. For a tool framed as a media editor, silent cloud transfer is a meaningful privacy and consent issue because users may assume local-only processing.

VirusTotal

65/65 vendors flagged this skill as clean.