tearsheet

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

• Third-party content exposure detected (medium risk: 0.30). The required workflow reads outsider-authored free text from GitHub-like external sources via WebSearch and “search the most recent 2 quarters of filings”/primary-source extraction, where arbitrary web pages or filings text could include attacker-injected instructions since the text is ingested at runtime as search results/content.