The Agent Skills Directory

[SAFE]: The skill provides a robust and well-documented framework for professional video production using standard web technologies.
[EXTERNAL_DOWNLOADS]: The skill references the GSAP library from a well-known CDN (jsDelivr) and describes the download of local AI models (Kokoro-82M) for text-to-speech functionality. These are standard resources for the skill's intended purpose.
[COMMAND_EXECUTION]: The skill includes Node.js utility scripts (animation-map.mjs and contrast-report.mjs) used for project validation. These scripts execute shell commands via npx to interact with the project's CLI and use Puppeteer for browser-based introspection.
[DYNAMIC_EXECUTION]: Diagnostic scripts utilize Puppeteer's evaluate function to measure DOM elements and GSAP timelines in a controlled browser environment.
[INDIRECT_PROMPT_INJECTION]: The skill identifies a potential attack surface through the ingestion of external transcript and audio data files. This is documented with clear guidance on quality checks and sanitization of transcript tokens (e.g., removing non-speech markers), and the data is primarily used to drive animation timing rather than logic.

hyperframes