aether

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly ingests public, user-generated live chat (YouTube/Twitch/Bilibili) — see references/chat-platforms.md and the Workflow/Core Contract which require reading live chat and feeding it into the LLM→TTS→avatar→OBS pipeline — so untrusted third-party messages are read and can directly influence agent outputs and downstream actions.