The Agent Skills Directory

[SAFE]: The skill is a documentation-focused component for implementing a specific visual effect. It does not contain any malicious code, external exfiltration, or obfuscation.
[INDIRECT_PROMPT_INJECTION]: The skill processes text from the DOM to create a split-text effect. While this involves ingesting external data, the implementation is secure.
Ingestion points: Text is retrieved from DOM elements matching [data-masked-reveal] via element.textContent in SKILL.md.
Boundary markers: Not applicable to this UI utility.
Capability inventory: The skill modifies the DOM using element.innerHTML to wrap words in spans.
Sanitization: The code includes a robust escapeHTML function that sanitizes word content before it is inserted into the DOM, preventing Cross-Site Scripting (XSS) or injection attacks.
[COMMAND_EXECUTION]: No shell commands or system-level execution patterns were detected.
[EXTERNAL_DOWNLOADS]: The skill references GSAP and ScrollTrigger, which are well-known and trusted front-end animation libraries.

masked-reveal