canva-brand-check

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

• Third-party content exposure detected (high risk: 0.75). The workflow calls Canva:list-brand-kits and Canva:get-design-thumbnail/Canva:get-design-content to ingest brand-kit and design text/imagery at runtime; these are outsider-authored (public/third-party Canva content and brand-kit assets not authored by the operating user) and become LLM-readable context via the connector payloads.