- Add separate isWaitingForResponse state to track loading
- Set to true before mutation, false only in onSuccess/onError
- Combine with mutation.isPending for reliable isLoading state
- Typing indicator now stays visible throughout entire AI request
- Add proper cache control headers in static.ts
- HTML files: no-cache to always get fresh version
- Assets: long cache with immutable (content-hashed filenames)
- Improve typing indicator animation (larger dots, faster animation)
- Add animated typing indicator (bouncing dots) when AI is processing
- Show "Thinking..." status in chat header while processing
- Add translations for AI chat (en/de): title, welcome, placeholder, thinking, error
- Auto-refresh tasks, labels, and user data after AI response
- Disable input field while AI is processing
- Fix translation keys to use ai.chat.* namespace
Paul Nothaft
paul