feat(agent): add voice input and photo attachments to assistant

Wire mic through Whisper-compatible transcriptions on LLM_BASE_URL. Photos upload to MinIO and reach the vision model as base64 image_url parts.
2026-07-05 02:01:48 -05:00
parent 876a283d47
commit c8db5475d3
20 changed files with 636 additions and 65 deletions
@@ -49,6 +49,8 @@ OPENPLANTBOOK_CLIENT_SECRET=

 # LLM assistant (OpenAI-compatible — Ollama, vLLM, LiteLLM, etc.)
 # Leave LLM_BASE_URL unset to use the built-in mock provider (CI / local without a model).
+# Voice input uses POST {LLM_BASE_URL}/audio/transcriptions (Whisper-compatible).
+# Photo messages use vision via the same chat/completions endpoint.
 LLM_PROVIDER=openai
 LLM_BASE_URL=
 LLM_API_KEY=