Ai Video Ai

• Outcome: Review
• Latest audit: May 29, 2026 at 9:34 PM UTC
• Version:
  1.0.0

Overview

This cloud video-editing skill largely matches its purpose, but it can automatically create remote sessions and route broad user requests to a third-party backend.

Install only if you are comfortable sending videos, prompts, and session metadata to NemoVideo's cloud service. Avoid sensitive or confidential footage unless you trust that provider, and prefer explicit confirmation before first connection, upload, or export.

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

Low

• Confidence: 91% confidence
• Finding: The skill instructs the agent to inspect local install paths and runtime metadata to derive attribution headers, which is not necessary for core video processing and expands access to local environment details. Even though the specified paths are limited, this creates unnecessary local-data inspection behavior and can normalize broader filesystem probing by a remote-service skill.

Vague Triggers

Medium

• Confidence: 88% confidence
• Finding: The starter prompts are extremely broad and overlap with ordinary video-editing requests, increasing the chance the skill activates when the user did not specifically intend to send content to this backend. In this context, accidental invocation is risky because the skill uploads user media and instructions to a third-party cloud service and may initiate authentication/session setup automatically.

Vague Triggers

Medium

• Confidence: 94% confidence
• Finding: The routing table includes an 'Everything else' catch-all that sends arbitrary editing requests to the SSE backend, lacking clear scope boundaries. This makes over-collection and unintended remote transmission more likely, especially when user requests are generic and not clearly directed at this specific service.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill explains cloud rendering mechanics but does not clearly foreground, at the point of use, that user media, prompts, and session data are uploaded to and processed by an external backend. For a media-processing skill handling potentially sensitive videos, insufficient disclosure materially raises privacy and consent risks.

VirusTotal

66/66 vendors flagged this skill as clean.