product-demo

Pass

Audited by Gen Agent Trust Hub on Aug 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill includes Node.js templates (mix-audio.mjs, retime-vo.mjs, record.mjs) that programmatically execute ffmpeg, ffprobe, and playwright via the child_process module. These tools are used for video encoding, audio mixing, and frame capture.
  • [EXTERNAL_DOWNLOADS]: The pipeline depends on playwright, which downloads browser binaries during setup. It also suggests adding external skills for Remotion and ElevenLabs to enhance the workflow. The skill references legitimate vendor resources including a playbook and a reference repository from the author's official GitHub account.
  • [PROMPT_INJECTION]: The skill ingests and renders untrusted application code and components from the target repository, creating a surface for indirect prompt injection.
  • Ingestion points: Local repository components, CSS files, and data fixtures.
  • Boundary markers: None identified in the rendering or processing scripts.
  • Capability inventory: File system access, local HTTP server instantiation, and shell command execution via ffmpeg and Playwright.
  • Sanitization: No explicit validation or escaping of repository-sourced content before rendering.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 26, 2026, 05:41 PM