regenerating-overviews

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

• Third-party content exposure detected (high risk: 0.85). Outsider-authored free text from release bodies inside the server response to releases admin overview inputs <slug>... is passed into the LLM as search_result content blocks in step 2 (i.e., r.content / media URLs originate from outside the operator and can contain prompt-injection strings).