v1.4.7
OpenCodeSummary
This release introduces support for Claude Opus 4.7 with xhigh adaptive reasoning and defaults to showing summarized thinking for this model. Several critical bugs were fixed, including issues with TUI pasting, agent overwriting, and provider error handling during session replay. Additionally, performance improvements were made to the bash tool and workspace synchronization.
New Features
- Enabled support for Claude Opus 4.7 with xhigh adaptive reasoning.
- Introduced showing summarized thinking by default for Claude Opus 4.7.
Bug Fixes
- Fixed requests failing on Cloudflare AI Gateway for OpenAI reasoning models because `max_tokens` was being dropped.
- Fixed requests requiring stored responses failing because Azure models did not default `store=true`.
- Fixed pasting files or large text in the TUI causing content to be inserted twice.
- Fixed the command line `--agent` argument being overwritten by the session's saved agent.
- Fixed TUI sync state breaking due to empty LSP, MCP, formatter, and session status responses.
- Fixed an issue where stale bot messages were not aligning with the actual 60-day threshold.
Improvements
- GitHub Copilot gpt-5-mini now uses low reasoning effort for better request compatibility.
- Provider sign-in context now carries across workspace sessions.
- TUI plugins now load against the correct project when multiple directories are open.
- The bash tool uses less memory when processing large command output.
- Experimental workspaces now wait for sync to finish before returning writes, reducing stale reads and missed updates.
- Session restore can now replay a session into another workspace in batches.
- Sessions now retry provider 5xx errors even when the provider SDK does not mark them as retryable.
- Desktop builds now display a Beta or Dev badge in the title bar when applicable.