monitor-herdr

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

• Third-party content exposure detected (medium risk: 0.30). Source: Herdr pane transcript text read via the runtime workflow tick → scripts/herdr_socket.py:read_pane_text using pane.read (source recent_unwrapped) and fed into the monitor’s classifier/prompt-builder for resuming stopped panes.