Image To Video On Canva

• Outcome: Pass
• Latest audit: May 29, 2026, 8:39 PM
• Version:
  1.0.0

Overview

This is a disclosed cloud video-generation skill, but users should know it sends prompts and selected media to NemoVideo and creates or uses a service token/session.

Install only if you are comfortable sending selected images, media URLs, edit prompts, render state, and a NemoVideo token/session identifier to the NemoVideo service. Ask the agent to confirm before uploads or generation when a request is ambiguous, and avoid private or regulated media unless you have reviewed the provider’s terms and retention practices.

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: 92% confidence
• Finding: The invocation examples are broad marketing-style phrases around common photo/video editing tasks, which increases the chance the skill activates on ordinary user requests without a clear, narrow trigger. That can unexpectedly route user images and prompts into this remote service, creating consent and data-flow risks even if the author likely intended convenience rather than abuse.

Vague Triggers

Medium

• Confidence: 96% confidence
• Finding: The routing table includes an 'Everything else' catch-all that sends most non-export/upload requests into the SSE/chat action. This is overly permissive and can cause unintended activation from general creative-editing language, making accidental remote processing of user content more likely.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill instructs the agent to generate anonymous tokens and immediately call external APIs, but does not present a prominent user-facing warning about credential creation, account/session linkage, or that content will be transmitted to a third-party service. This creates a transparency and consent problem, particularly because the flow starts automatically on first interaction.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: Although the text mentions server-side rendering, it does not clearly foreground a privacy warning that uploaded images and edit instructions are processed on remote GPU infrastructure operated by an external service. Users may reasonably interpret this as a convenience feature without appreciating the data-handling implications.

VirusTotal

65/65 vendors flagged this skill as clean.