Skills
skills/gooseworks-ai/gooseworks-ads-skills/create-programmatic-motion-video-ad/Gen Agent Trust Hub

create-programmatic-motion-video-ad

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes system commands via ffmpeg for video assembly, ffprobe for metadata verification, and volumedetect for audio analysis. These are necessary for the programmatic creation of MP4 files.
  • [EXTERNAL_DOWNLOADS]: Fetches brand assets from user-provided URLs and interacts with the ElevenLabs API to generate audio content. These external interactions are essential for the skill's primary function of ad creation.
  • [PROMPT_INJECTION]: Potential for indirect prompt injection through ingested external content: 1. Ingestion points: Product briefs, brand websites, and asset folders mentioned in SKILL.md. 2. Boundary markers: No specific delimiters are defined in the instructions to isolate untrusted data. 3. Capability inventory: Use of playwright for dynamic HTML rendering and ffmpeg for file processing. 4. Sanitization: The instructions do not specify any sanitization or validation logic for data retrieved from external sources.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 03:07 PM
Security Audit — agent-trust-hub — create-programmatic-motion-video-ad