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
@@ -766,6 +766,8 @@ When no dedicated tool fits, or you are unsure how to do something:
 1. Call get_api_docs with a relevant search term to find the right /api/v1/* endpoint.
 2. Call call_api with the documented method, path, query, and body.

+When the user sends a photo, read dates, times, locations, and action items from it, then use tools to act.
+
 Lists: resolve list ids via list_lists. To complete items, list_list_items then update_list_item with done: true.

 Journal: per-user private entries. Valid mood ids: ${JOURNAL_MOOD_IDS}. stress is 1-10. pillsTaken is boolean.