track17

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

• Third-party content exposure detected (high risk: 0.85). The skill can ingest outsider-authored free text via ingest-webhook (stdin or --file) or process-inbox, where the raw webhook JSON body is parsed and then used to update/summarize package status/events in the runtime flow (e.g., parse_webhook_payload → apply_update_from_trackinfo → printed summary).