Scope: create a design plan for supporting audio and video as multimodal context, following the existing image-context pattern.\n\nAcceptance criteria:\n- Document the proposed config/API shape for audio and video context.\n- Clarify that context values are URL or base64 only.\n- Call out provider translation boundaries and unsupported-provider behavior.\n- Include a focused implementation and test plan.\n\nOut of scope: implementation changes, local path handling, file uploads, or audio/video generation columns.
Scope: create a design plan for supporting audio and video as multimodal context, following the existing image-context pattern.\n\nAcceptance criteria:\n- Document the proposed config/API shape for audio and video context.\n- Clarify that context values are URL or base64 only.\n- Call out provider translation boundaries and unsupported-provider behavior.\n- Include a focused implementation and test plan.\n\nOut of scope: implementation changes, local path handling, file uploads, or audio/video generation columns.