The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references and downloads libraries and assets from well-known services.\n
Fetches Google Fonts from fonts.googleapis.com.\n
Loads script libraries (GSAP, Anime.js, Three.js, Iconify) from official CDNs including cdnjs.cloudflare.com, cdn.tailwindcss.com, and code.iconify.design.\n
References background images from Unsplash (images.unsplash.com).\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted user-supplied data to generate marketing copy and page structures, which creates a potential surface for indirect prompt injection.\n
Ingestion points: Data enters the agent context through fields defined in references/01-intake-and-mode-selection.md, such as brand_theme, value_props, and primary_cta.\n
Boundary markers: The skill instructions do not specify explicit delimiters or "ignore instructions" warnings for the data being interpolated into the generated output templates.\n
Capability inventory: The skill allows the agent to write files (HTML, JS, CSS) to the workspace and execute included Python scripts (validate_intake.py and build_animation_manifest.py) to process this data.\n
Sanitization: While the skill follows standard implementation patterns (e.g., using React for rendering or predefined HTML skeletons), there is no specific sanitization step mentioned for checking the semantic content of user-provided strings for embedded instructions.

landing-page-studio