You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Real-time Voice Transcription - Voice input now shows live transcript updates as you speak with enhanced visual feedback
Improved Voice Architecture - Upgraded to react-speech-recognition library for more reliable speech processing and better integration
🚀 Improvements & UI Changes
Enhanced Code Navigation - Added scroll buttons for long code blocks with smooth scrolling to top/bottom functionality
Smart UI Controls - Scroll buttons appear only when needed and when hovering over messages
Model Context Protocol Update - Updated to version 0.3.0-preview.2 for enhanced MCP compatibility
AI Provider Token Handling - Improved max-tokens parameter management for better response generation
🐛 Bug Fixes
Enhanced Tool Call Reliability - Fixed tool persistence across chat sessions for OpenAI, Claude, and Gemini, reducing AI confusion when resuming conversations
