why

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

• Third-party content exposure detected (medium risk: 0.30). Source control, PR, issue/ticket tracker, long-form docs, chat, observability, error tracking, and analytics evidence are all fetched/read at runtime by the investigators using enabled MCPs (Step 3 “queries all seven in parallel”), which can include outsider-authored free text from non-first-party users (e.g., issue comments, chat messages, Sentry event text, ticket descriptions/attachments).