muapi-floor-plan-rendering

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

• Third-party content exposure detected (high risk: 0.80). The skill's SKILL.md explicitly accepts a base_plan_image (image_url) in the Inputs and uses that reference image in Phase A/Phase B conversion steps, so the agent will fetch and interpret arbitrary external images (potentially untrusted/user-generated) as part of its workflow and those assets can materially influence generation behavior.