angle-writer

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow ingests user-provided “brand kit”/“PASTE BRAND READER OUTPUT” text and the user’s freeform instruction text (“Write 6 ad angles…”) to generate ad angles, so outsider-authored text directly enters the LLM context without selecting a specific trusted item first.