debugger

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow for debugger consumes outsider-provided free text directly via the user-driven /debugger triggers and, in walkthrough/converse mode, passes “spoken barge-in utterances” as the next spoken command/question to the LLM-backed /ask handler—so an outsider can inject instructions through natural language without selecting a specific item first.