remix-chatgpt-ad-from-sample

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

• Third-party content exposure detected (high risk: 0.75). Yes—outsider-authored free text from the public/third-party “source sample” is fetched at runtime via get_ad_template { template_id: source_sample_id } and its recipe.conversation / extracted_script (which contains the original user question and assistant answer) is then used to derive the in-context conversation that is shown/approved and passed into the renderer/LLM workflow.