Whiteboard Video Workflow

• Outcome: Review
• Latest audit: May 29, 2026, 9:31 PM
• Version:
  0.1.0

Overview

This skill fits its whiteboard-video purpose, but it automatically runs dependency setup from another skill and sends subtitle-derived prompts to RunningHub, so users should review it before installing.

Install only if you trust this skill and the separate whiteboard-animation skill it invokes. Prefer running dependency setup in a sandbox or virtual environment, use a revocable RunningHub API key, keep output paths inside your workspace, and avoid confidential subtitles unless third-party image processing is acceptable.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Behavioral AST: exec() Call, eval() Call, Dynamic Import
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (3)

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The environment precheck explicitly says the script will automatically detect and install fixable dependencies, meaning the workflow may modify the local Python environment before obtaining clear user consent. Automatic package installation through a shell-invoked script expands the attack surface and can unexpectedly change system state, especially if dependency sources, versions, or install scope are not constrained and disclosed.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The skill directs the agent to use an external API key, transmit user-provided prompts to a third-party image generation service, and download/write files locally, but it does not require explicit user consent or a warning about those side effects. In an agent context, this can cause unintended disclosure of sensitive prompt content and unexpected filesystem changes, especially when the user may believe the operation is local-only.

Missing User Warnings

Medium

• Confidence: 81% confidence
• Finding: The skill instructs the agent to write files under a user-controlled projectRoot and later execute a local Python script, but the documentation does not explicitly surface these filesystem and subprocess side effects. In an agent setting, lack of clear side-effect disclosure increases the risk that users or orchestrators invoke the skill without understanding it will modify disk state and launch code, which can amplify damage if inputs point to sensitive locations or if downstream execution is unexpectedly trusted.

VirusTotal

66/66 vendors flagged this skill as clean.