Online Free Ai Tool

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:41 PM UTC
• Version:
  1.0.0

Overview

This is a cloud video-editing skill that uses a NemoVideo token and uploads user-selected media to an external rendering API, which is disclosed and aligned with its purpose.

Install only if you are comfortable sending selected videos, images, audio, and edit prompts to NemoVideo's cloud service. Use a dedicated NEMO_TOKEN for this service, watch for credit or plan usage, and ask the agent to confirm before uploads or exports when working with sensitive media.

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
• 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 (7)

Description-Behavior Mismatch

Medium

• Confidence: 94% confidence
• Finding: The manifest presents the skill as a simple video editor, but the documented behavior also performs authentication, anonymous token acquisition, session creation, and credit/account handling against a remote backend. This is a scope-expansion and transparency problem: users may invoke a media-editing skill without realizing it will contact external services and manage credentials/tokens on their behalf.

Description-Behavior Mismatch

Medium

• Confidence: 89% confidence
• Finding: The skill is advertised primarily as a video-clip editing tool, but its accepted inputs include images and audio files beyond the stated video formats. This mismatch weakens informed consent and can cause users to provide additional media types to a cloud service without clear upfront disclosure.

Vague Triggers

Medium

• Confidence: 86% confidence
• Finding: The suggested invocation language is broad enough that ordinary conversation about editing or exporting could trigger the skill unintentionally. Because this skill initiates backend connections and may upload/process user media remotely, accidental activation can lead to unintended data handling and external API use.

Vague Triggers

Medium

• Confidence: 93% confidence
• Finding: The catch-all routing rule sends 'everything else' to the SSE editing path, creating ambiguous trigger boundaries. In a skill that can contact remote services and act on uploaded media, such permissive routing increases the chance of unintended processing from loosely related user prompts.

Missing User Warnings

Medium

• Confidence: 97% confidence
• Finding: The skill description and getting-started flow do not clearly warn users that uploaded media and prompts are sent to a cloud backend for processing. For a media-editing skill, this is significant because video, audio, and images may contain sensitive or personal information, and users may reasonably expect local-only handling unless told otherwise.

Missing User Warnings

Low

• Confidence: 84% confidence
• Finding: The metadata indicates use of environment credentials and local config paths, but the markdown does not clearly warn users that the skill may access `NEMO_TOKEN` and configuration under `~/.config/nemovideo/`. Lack of disclosure about credential and local-config usage undermines informed consent and can surprise users about the skill's access scope.

Natural-Language Policy Violations

Medium

• Confidence: 72% confidence
• Finding: The session creation flow hard-codes `language":"en"` without asking the user or respecting their locale. This is primarily a quality, accessibility, and transparency issue rather than a direct security exploit, but it can cause misprocessing of user content and reduce user control over how data is handled by the remote service.

VirusTotal

64/64 vendors flagged this skill as clean.