Editor Ai Mac

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

Overview

This video-editing skill appears legitimate, but it can automatically contact a remote service and route broad or ambiguous requests there without clear user confirmation.

Review before installing. Use it only if you are comfortable with your video files, editing prompts, timeline state, and render jobs being sent to Nemo Video's remote backend. Avoid confidential, workplace, or sensitive personal footage unless you understand the provider's retention and privacy terms, and require explicit confirmation before the skill creates a session, uploads files, or starts 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 (5)

Vague Triggers

Medium

• Confidence: 88% confidence
• Finding: The skill advertises very broad trigger phrases like "edit my video clips" and generic export-oriented wording, which can cause invocation on loosely related requests. In an agent setting, overbroad activation increases the chance that user content is sent to the remote backend or that backend sessions are created when the user did not clearly intend to use this specific third-party video-editing service.

Vague Triggers

Medium

• Confidence: 93% confidence
• Finding: The routing table includes an "Everything else" branch that sends unmatched requests into the SSE editing path. This catch-all behavior is dangerous because ambiguous or unrelated user prompts may still be forwarded to the backend, causing unintended data disclosure, unwanted actions, or surprise consumption of credits.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: Although the text mentions that rendering happens server-side, it does not present a clear upfront warning that uploaded media and editing instructions are transmitted to a third-party remote service for processing. For a media skill handling potentially sensitive videos, this can mislead users about privacy expectations and result in unintentional disclosure of personal or confidential content.

Missing User Warnings

Low

• Confidence: 84% confidence
• Finding: The skill instructs the agent to automatically obtain an anonymous token from the backend when no local token is present, without clearly notifying the user. This reduces transparency around authentication and backend account creation, and can surprise users by initiating third-party service usage and credit consumption without explicit consent.

Natural-Language Policy Violations

Medium

• Confidence: 78% confidence
• Finding: Hard-coding `"language":"en"` during session creation is primarily a safety and UX issue rather than a direct exploit primitive, but it can lead to misinterpretation of user instructions and incorrect edits for non-English users. In this context, the risk is limited, yet forcing a language without user choice can cause unintended processing outcomes and weaken informed consent for multilingual interactions.

VirusTotal

66/66 vendors flagged this skill as clean.