hyperframes-registry

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.65). The hyperframes add <name> workflow installs registry items by fetching from the configured public registry URL (e.g., hyperframes.json#registry points to a raw GitHub URL) and then the installed block/component HTML (outsider-authored) is read/loaded into the runtime DOM/JS context via data-composition-src (blocks) or pasted snippet content (components), creating an indirect prompt-injection surface if that HTML/JS contains attacker-controlled free text.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The templates and demo HTML include CDN script tags that are fetched and executed at runtime (notably https://cdn.jsdelivr.net/npm/gsap@3.14.2/dist/gsap.min.js and https://cdn.jsdelivr.net/npm/three@0.147.0/build/three.min.js), and the skill’s demo/templates rely on those remote JS files to run timelines/VFX, so they are runtime external dependencies that execute code.