import

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

• Third-party content exposure detected (high risk: 0.70). The workflow ingests outsider-authored free text from a user-provided mem0 export file (Step 1/2: the runtime parses <path-to-export-file> via parse_export_file.py, then Step 4 passes record.content into add_memory as text=...), and that export content is not guaranteed to be authored by the operating user.