why

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

• Third-party content exposure detected (high risk: 0.75). The required workflow explicitly instructs investigators to fetch and read “issue/ticket tracker”, “long-form documents”, and “real-time team chat” contents (e.g., PR bodies, ticket descriptions, Slack threads) at runtime, which are typically outsider-authored free text and then passed into the synthesizer context as “investigator findings,” enabling indirect prompt injection.