Social Copy Generator Free

• Outcome: Pass
• Latest audit: May 29, 2026, 8:40 PM
• Version:
  1.0.0

Overview

This is a disclosed cloud video and social-copy generator; videos and prompts are sent to NemoVideo, but the artifacts do not show hidden, destructive, or unrelated behavior.

Install only if you are comfortable sending selected videos, URLs, prompts, and render drafts to NemoVideo for cloud processing. Avoid confidential or sensitive media, keep NEMO_TOKEN private, and invoke it only for explicit social video, caption, or export tasks.

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: 89% confidence
• Finding: The manifest presents the skill as narrow social-copy generation, but the body describes a much broader remote video editing, rendering, session, upload, and export pipeline. This mismatch can mislead users and downstream systems about the skill's true capabilities and data flows, reducing informed consent and making overbroad invocation or use more likely.

Context-Inappropriate Capability

Medium

• Confidence: 85% confidence
• Finding: The skill allows arbitrary remote media ingestion by URL, which expands the trust boundary beyond user-supplied local files and is not clearly necessary for the advertised function. This can enable fetching unexpected third-party content, increase privacy and compliance risk, and potentially be abused to process untrusted or sensitive media without clear user awareness.

Vague Triggers

Medium

• Confidence: 84% confidence
• Finding: The invocation examples use broad phrases that are common in normal editing or content requests, increasing the chance the skill activates outside a user's clear intent. Overbroad activation is risky here because the skill's first-step behavior includes connecting to external APIs and potentially initiating token/session setup before the user has meaningfully consented.

Vague Triggers

Medium

• Confidence: 91% confidence
• Finding: The catch-all routing rule sends 'Everything else' to the SSE workflow, making activation scope ambiguous and effectively defaulting many unrelated prompts into the remote processing path. In this context, ambiguity is more dangerous because SSE drives backend actions tied to sessions, state, uploads, and rendering on an external service.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The description does not clearly warn users that uploaded video content is transmitted to a remote processing service, even though the skill performs server-side uploads, session creation, and cloud rendering. This undermines informed consent for potentially sensitive media and is especially concerning because video files may contain faces, voices, locations, and other personal or confidential information.

VirusTotal

66/66 vendors flagged this skill as clean.