ai-amazon-storefront-design

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

• Third-party content exposure detected (medium risk: 0.65). Skill runtime context composition can ingest user-supplied materialUrl/materialText/materialFile; if materialUrl points to public web pages or materialText/files contain outsider-authored free text, that text is read and fed into the LLM via the skill’s contextComposer path (materialUrl/materialText/materialFile → LLM context), creating indirect prompt-injection risk.