Starstruk Logo MarkStarstruk

cli-v3.0.38

kubernetes-sigs/agent-sandboxcli-v3.0.38Jul 7, 2026by github-actions[bot]

AI Summary

This release introduces a comprehensive UI redesign with a new blue and amber accent palette, restyled chat interfaces, and fixes for file reading and model context handling.

Key Highlights

  • New plan/act accent palette (blue/amber) applied across dialogs, model selector, and markdown
  • Restyled chat input with a minimal frame and bold accent glyph instead of tinted backgrounds
  • Markdown accents are now tinted by the mode (plan/act) they were produced in
  • Models in the live catalog without context windows now default to a 128K input-token limit
  • read_files now tolerates malformed input from weaker models

New Features

  • New accent palette for light-theme variants tuned for contrast
  • Default thinking level picker set to Medium
  • Tencent TokenHub provider support (from SDK v0.0.55)
  • Models without context windows get a usable 128K context budget

Full Release Notes


- New plan/act accent palette: act mode is now blue (`#79b8ff`) and plan mode amber, replacing the old cyan/yellow — applied across dialogs, the model selector, config, onboarding, markdown, and syntax highlighting, with light-theme variants tuned for contrast
- Restyled chat input: a minimal frame with full-width horizontal rules and a bold accent prompt glyph instead of the tinted background, plus slimmer user-message bubbles
- Assistant markdown accents are now tinted by the mode (plan/act) they were produced in
- Polished the status bar usage display and ClinePass model name
- Harmonized the success/diff green and dark syntax-highlighting colors with the new brand palette
- The thinking-level picker now defaults its cursor to Medium instead of Off
- `read_files` now tolerates malformed input from weaker models: line-range entries (`start_line`/`end_line`) sent as separate array items are coalesced back onto the preceding file path instead of being rejected (from SDK v0.0.58)
- Models in the live catalog that don't report a context window now default to a 128K input-token limit, so under-specified models get a usable context budget (from SDK v0.0.57)

**Full Changelog**: https://github.com/cline/cline/compare/cli-v3.0.37...cli-v3.0.38