Image To Video Local Ai

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

Overview

This skill appears to be a cloud image-to-video connector, but its “Local AI” branding and broad automatic routing could lead users to send images or prompts to NemoVideo without clear intent.

Install only if you are comfortable sending selected images, prompts, and render/session data to NemoVideo cloud services. Before using it, confirm that the agent will ask before creating tokens, uploading files, or sending free-form edit prompts, and avoid sensitive photos unless you trust the provider’s privacy 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 (3)

Vague Triggers

Medium

• Confidence: 95% confidence
• Finding: The onboarding prompts and invocation examples are broad enough that ordinary phrases like "export 1080p MP4" or "convert my still images" could trigger the skill outside a clearly scoped user intent. Because this skill uploads user content and communicates with a third-party cloud backend, accidental activation can cause unintended transmission of files or prompts and unexpected account or credit usage.

Vague Triggers

Medium

• Confidence: 97% confidence
• Finding: The catch-all rule routes "Everything else" to the SSE action, creating an overly permissive trigger surface for arbitrary prompts. In this skill, SSE sends user text to a remote backend that can drive edits and stateful operations, so ambiguous matching increases the risk of unintended external data disclosure and unreviewed remote actions.

Missing User Warnings

Medium

• Confidence: 99% confidence
• Finding: The user-facing description implies "local AI" while the implementation actually sends images, prompts, and session data to a cloud processing backend. This is a material transparency failure: users may reasonably believe their media stays local, when in fact sensitive files and instructions are transmitted to external infrastructure and authenticated with a service token.

VirusTotal

64/64 vendors flagged this skill as clean.