- VSCode/Files: added file stat support in the extension bridge (thanks to @geekifan).
- Chat/Models: added arrow-key navigation for thinking-mode selection in model controls (thanks to @daveotero).
- Files: added HTML preview support in the file viewer (thanks to @nguyenngothuong).
- Chat/JSON: added an interactive JSON tree viewer with collapse/expand controls and richer color cues for easier inspection of large structured outputs (thanks to @nguyenngothuong).
### Improvements
- Security/Chat: user messages now escape raw HTML by default (thanks to @kalac2232).
- Desktop/Performance: reduced Tauri shell CPU/GPU overhead during longer sessions.
- Sessions/Drafts: draft chat config now stays synced with the selected draft target directory.
- Chat: improved error message readability with clearer styling and safer word-wrapping (thanks to @nguyenngothuong).
- Desktop/Settings: window vibrancy now correctly controls macOS window transparency, and settings copy now clarifies when full transparency changes take effect.
- Usage/Providers: added ZhipuAI quota tracking and fixed MiniMax coding-plan and GitHub Copilot overusage calculations (thanks to @kalac2232, @baruchvitorino, @ebrainte).
## VS Code
### New
- Files/Markdown: added filesystem stat support in the extension bridge to validate markdown targets reliably before file handling flows (thanks to @geekifan).
- Chat/Models: added arrow-key navigation for thinking-mode selection in model controls (thanks to @daveotero).
- Chat/JSON: added an interactive JSON tree viewer with collapse/expand controls and richer color cues for easier inspection of large structured outputs (thanks to @nguyenngothuong).
### Improvements
- Security/Chat: user messages now escape raw HTML by default (thanks to @kalac2232).
- Sessions/Drafts: draft chat config now stays aligned with the active draft target directory.
- Chat: improved error message readability with clearer styling and safer word-wrapping (thanks to @nguyenngothuong).
- Reliability/Streaming: proxy handling now normalizes identity encoding, strips conflicting compression headers and hop-by-hop response headers, and suppresses expected upstream SSE close errors to reduce noisy disconnect failures (thanks to @jwcrystal, @Jovines, @JiwaniZakir, @shekohex).
### Fixes
- Usage: added ZhipuAI quota tracking and fixed MiniMax coding-plan plus GitHub Copilot overusage calculations (thanks to @kalac2232, @baruchvitorino, @ebrainte).