brand-launch-video-star

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

• Third-party content exposure detected (medium risk: 0.65). The skill’s required workflow says it will “Browse current official sources and extract” brand/product details (SKILL.md:74-82), which implies fetching public/outsider-authored web pages at runtime and then feeding that extracted free text into the agent’s LLM planning context, creating an indirect prompt-injection surface.