synthesis-meeting-transcripts

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly searches the user's Gmail and Google Drive for Gemini-generated meeting docs and fetches their full content (see SKILL.md Step 2–3 and optional-workspace-mcp/fetch-meeting.py which calls search_drive_files and get_drive_file_content), and that untrusted/user-generated document content is parsed and used to choose files, extract notes/actions, and drive saves/commits—so third-party content can directly influence agent actions.