meeting-analyzer

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

• Third-party content exposure detected (medium risk: 0.30). Outsider-authored meeting notes are read from a user-provided --input file and parsed at runtime by scripts/meeting_notes_parser.py via read_notes(path)/classify(...), making any free text in those notes (including poison content) directly ingestible by the workflow.