Free Editor Text

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

Overview

This video text-editing skill is not clearly malicious, but it needs review because it can automatically connect to a cloud service and send broad prompts or user media without strong user confirmation.

Install only if you are comfortable sending videos, audio, images, prompts, and related metadata to NemoVideo's cloud service. Use non-sensitive media, protect or rotate any NEMO_TOKEN, and ask the agent to confirm before uploading files, sending broad edit prompts, or starting exports.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• 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
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (4)

Vague Triggers

Medium

• Confidence: 86% confidence
• Finding: The invocation language is broad enough that ordinary conversation could trigger the skill unexpectedly, causing uploads or editing actions to be initiated without clear user intent. In this skill, unintended activation is more concerning because it connects to a remote backend and may send user media or consume credits.

Vague Triggers

Medium

• Confidence: 89% confidence
• Finding: The catch-all routing rule sends 'everything else' to the SSE editing path, which makes accidental or ambiguous prompts likely to trigger remote processing. Because the SSE path can issue backend edits and state changes, broad routing increases the risk of unintended cloud actions, credit consumption, and confusing behavior.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill encourages users to upload media and states that processing happens on cloud GPUs, but it does not present a clear, prominent warning that files are transmitted to third-party remote services. For a media-handling skill, this omission materially affects privacy expectations and may lead users to disclose sensitive video or audio content without informed consent.

Natural-Language Policy Violations

Medium

• Confidence: 78% confidence
• Finding: Hard-coding the session language to English without user choice can cause prompts or content to be processed under the wrong language setting, which may reduce accuracy and mishandle user data semantics. While not a severe security bug, it is a trust and consent issue because the skill silently fixes a user-affecting backend parameter.

VirusTotal

66/66 vendors flagged this skill as clean.