The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill downloads the hyperframes package from the npm registry and the gsap animation library from the jsdelivr CDN.
Evidence: npx --yes hyperframes@0.8.206 in package.json and README.md.
Evidence: <script src="https://cdn.jsdelivr.net/npm/gsap@3.14.2/dist/gsap.min.js"></script> in examples/demo/index.html.
[REMOTE_CODE_EXECUTION]: The skill executes the hyperframes CLI tool via npx to perform video checking and rendering tasks.
Evidence: npx --yes hyperframes@0.8.206 check and render commands in package.json and README.md.
[COMMAND_EXECUTION]: Shell commands are used to invoke the rendering pipeline and manage the development environment.
Evidence: Scripts in examples/demo/package.json use npx to run HyperFrames commands.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied data that is rendered directly into video content, creating a potential surface for indirect prompt injection.
Ingestion points: User-provided strings for ctaLabel, folName, folHandle, revQuote, and revBy are accepted as composition variables in cta.html, follow.html, and review.html.
Boundary markers: There are no explicit delimiters or safety instructions provided to the agent or user regarding the content of these strings.
Capability inventory: The npx hyperframes tool generates video files (MP4, WebM) incorporating these inputs across various scripts.
Sanitization: The provided code shows the inputs being placed directly into the DOM (e.g., via data-var-text or as text content) without visible escaping or validation.

social-overlays