The Web Speech API hook was sending each recognition result as a
separate fragment via onTranscriptUpdate, causing the transcription
shape to display fragmented/echoed conversation. Now the hook
accumulates the full transcript internally and always sends the
complete text, so the shape receives one unified conversation.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Add useMaximize hook to all shapes using StandardizedToolWrapper:
- MapShapeUtil, MultmuxShapeUtil, MarkdownShapeUtil
- ObsNoteShapeUtil, ImageGenShapeUtil, VideoGenShapeUtil
- HolonShapeUtil, PromptShapeUtil, EmbedShapeUtil
- FathomMeetingsBrowserShapeUtil, FathomNoteShapeUtil
- HolonBrowserShapeUtil, ObsidianBrowserShapeUtil
- TranscriptionShapeUtil, VideoChatShapeUtil
All tools now have maximize/fullscreen functionality via the
standardized header bar.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>