coauthoring-docs

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

• Third-party content exposure detected (medium risk: 0.30). The skill ingests outsider-authored free text at runtime via user-provided “info dump” content and optionally reads linked/shared documents or current shared-document state when the user supplies those links/files (i.e., attacker-controlled text can be fed into Stage 1 context gathering and then used throughout drafting/testing).