marquee-loop

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s required runtime workflow reads a provided HTML/markup source (the sandboxed Neuform demo content embedded into demo/index.html via the runtime variable encodedHtml and then assigned to frame.srcdoc), which can include outsider-authored prompt text if that text is supplied into the skill that generated the HTML.