Free Generator No

• Outcome: Pass
• Latest audit: May 29, 2026, 8:45 PM
• Version:
  1.0.0

Overview

This is a cloud video-generation skill that discloses its NemoVideo API use and token/session flow, with no local executable code or destructive behavior found.

Install only if you are comfortable sending prompts, scripts, uploaded media, and media URLs to NemoVideo for cloud processing. Keep NEMO_TOKEN private, avoid sensitive or regulated content, and ask the agent to confirm before uploads or exports if credit use or privacy matters.

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
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands

Findings (5)

Description-Behavior Mismatch

Medium

• Confidence: 95% confidence
• Finding: The manifest frames the skill as simple text-to-video generation, but the body grants broader powers including uploads, authenticated session management, state inspection, credits queries, and export orchestration. This mismatch can mislead users and host platforms about the real data flows and permissions involved, increasing the chance of unintended invocation and unconsented third-party processing.

Context-Inappropriate Capability

Medium

• Confidence: 96% confidence
• Finding: The skill instructs the agent to automatically obtain anonymous auth tokens and create authenticated cloud sessions before doing anything else. That enables external account-like access and remote processing without an explicit user consent step, and it can be abused to send prompts or files to a third-party service under a silently provisioned token.

Vague Triggers

Medium

• Confidence: 90% confidence
• Finding: The activation examples use broad, ordinary language that could match many general creative requests, making the skill easier to invoke outside a clearly informed context. Because the skill then performs remote authentication and cloud processing, overbroad triggers raise the risk of accidental data transfer to the external service.

Vague Triggers

Medium

• Confidence: 94% confidence
• Finding: The catch-all routing rule sends 'everything else' to the SSE/chat action, creating an ambiguous and overly broad trigger surface. In this skill, that means many unrelated or insufficiently specific user inputs may be forwarded to the backend, increasing the chance of unintended prompt disclosure and unexpected remote actions.

Missing User Warnings

Medium

• Confidence: 98% confidence
• Finding: The skill lacks a clear warning that prompts, uploaded media, and session metadata are transmitted to a third-party cloud API. This is especially risky because the documentation encourages automatic setup and file upload workflows, so users may unknowingly expose sensitive text, media, or identifiers to an external service.

VirusTotal

64/64 vendors flagged this skill as clean.