Pet Adoption Video

• Outcome: Pass
• Latest audit: May 29, 2026, 8:31 PM
• Version:
  1.0.4

Overview

This is a cloud video-creation skill that asks for a NemoVideo token and sends video-project work to a remote backend, but those behaviors are disclosed in the artifact and match its video-rendering purpose.

Install only if you are comfortable using NemoVideo cloud processing for adoption-video work. Protect NEMO_TOKEN, review the service before sharing non-public shelter media, and upload only photos, videos, URLs, and pet details you are authorized to send to a third-party processor.

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
• 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 (4)

Description-Behavior Mismatch

High

• Confidence: 96% confidence
• Finding: The manifest markets the skill as a creative scripting and captioning assistant, but the body instructs the agent to authenticate to a remote backend, create sessions, upload files, render media, and export outputs. This scope mismatch is dangerous because users and hosting platforms may grant trust appropriate for a low-risk writing tool while the skill actually performs networked media processing and data transfer.

Context-Inappropriate Capability

Medium

• Confidence: 92% confidence
• Finding: The skill requests access to an environment token and local config path, then uses them for remote account/session operations that exceed the stated caption-writing purpose. This expands the trust boundary unnecessarily and can expose credentials or account-linked activity to a third-party service without clear necessity or informed consent.

Description-Behavior Mismatch

High

• Confidence: 97% confidence
• Finding: The documentation describes a full cloud-rendered video pipeline with session management, SSE messaging, uploads, state queries, and export/download behavior, which materially exceeds the manifest's creative-aid positioning. This can mislead users into providing media and sensitive descriptive data under false assumptions about local-only assistance, increasing privacy and supply-chain risk.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill instructs automatic connection, token acquisition, and session creation before doing anything else, but does not clearly warn users that their prompts, pet details, and uploaded media will be sent to an external service. Silent transmission of user content to a third party is a meaningful privacy and consent problem, especially when uploaded media may include people, locations, or shelter-identifying details.

VirusTotal

64/64 vendors flagged this skill as clean.