speech

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

• Third-party content exposure detected (high risk: 0.75). Outsider free text can be ingested into the LLM context via the user-provided --input/--input-file (or JSONL input/text/prompt) which is sent as the input field to the OpenAI TTS endpoint; this text is not sourced from the skill’s bundled references and may be authored by an outsider.