kinetic-typography

Pass

Audited by Gen Agent Trust Hub on Oct 25, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it transforms user-provided scripts into video animations.
  • Ingestion points: User text is incorporated into the PHRASES array within the generated HTML template in SKILL.md.
  • Boundary markers: The instructions mandate a planning phase (the "Brief") which helps structure the input, though it does not provide hard delimiters for untrusted data.
  • Capability inventory: The skill environment allows execution of npx hyperframes, ffmpeg, and ffprobe (documented in the Setup and Validate sections).
  • Sanitization: The provided HTML/JavaScript template correctly uses textContent to populate the DOM, which prevents basic cross-site scripting (XSS) risks in the rendered output.
  • [EXTERNAL_DOWNLOADS]: The skill fetches necessary assets from well-known services.
  • Downloads the GSAP (GreenSock Animation Platform) library from the jsDelivr CDN.
  • Retrieves font files from Google Fonts (fonts.googleapis.com).
  • [COMMAND_EXECUTION]: The skill relies on shell commands for its core functionality.
  • Executes npx hyperframes for project initialization, audio transcription, and music beat analysis.
  • Utilizes ffmpeg and ffprobe for video rendering and media stream validation.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 25, 2026, 12:48 AM