Aether

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

• Third-party content exposure detected (high risk: 0.90). The skill directly ingests public, user-generated live chat from YouTube/Twitch (see references/chat-platforms.md and the SKILL.md workflow "Chat Listener -> Message Queue -> LLM Engine") and uses those messages (including commands like !emotion / !ask) to drive LLM responses, TTS playback, avatar expressions, and scene/automation decisions, so untrusted third‑party content can materially influence agent behavior.