Cycle reasoning effort with Shift+Tab and show it on the prompt - #484
Merged
TheGreatAxios merged 2 commits intoAug 19, 2026
Merged
Conversation
Shift+Tab walks the current model's supported effort ladder (wrapping), rebuilds inference sources so the next turn picks it up, and refreshes the prompt border label. Plain Tab still toggles focus. Docs no longer claim Shift+Tab toggles auto mode.
TheGreatAxios
commented
Aug 19, 2026
TheGreatAxios
left a comment
Collaborator
Author
There was a problem hiding this comment.
Reviewed as reviewer of record against CL-6298.
Verdict: approve (GitHub will not let this identity formally approve its own PR).
Shift+Tab cycles the live model's supported effort ladder, rebuilds inference sources so the next turn picks it up, and the prompt border label plus a status flash show the new value. Plain Tab still toggles focus. Overlays still treat unshifted Tab as complete.
Checked:
- cycleReasoningEffort wraps the per-model ladder; undefined/unsupported current starts at the first supported level; non-reasoning models return undefined and the runner flashes instead of mutating config.
- Shell onKey handles both tab+shift and backtab, after overlay handling, and does not steal unshifted Tab.
- Runner updates config.reasoningEffort, calls buildSessionSources(), agentProxy.setSources, setPromptModelLabel, and flashes the new level.
- Live keybinding probe in keybindings.test.ts asserts the handler fires and focus is unchanged.
- README / config comments no longer claim Shift+Tab toggles auto.
Nits (not blocking):
- docs/TUI.md is the normative TUI spec and still documents Tab without Shift+Tab. Help overlay is updated via keybindings.ts; TUI.md should match.
- docs/PRODUCT.md does not mention the new binding (README does).
- First press when session effort is unset jumps to supported[0] (e.g. minimal on gpt-5), not next-after-the-implicit-role-default.
- /model does not clamp a previously cycled effort onto the new model's ladder. Pre-existing, but Shift+Tab makes edge values (none, xhigh) more reachable.
Commit message is plain English and matches the change.
PRODUCT and TUI specs now name the binding; implementation notes how the runner rebuilds sources and the prompt-border label.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Rebinds Shift+Tab to cycle reasoning effort for the live model and keeps the current effort visible on the prompt border. Plain Tab still toggles prompt/transcript focus.
Closes CL-6298.
Changes
cycleReasoningEffortwalks the model's supported ladder and wrapsTest plan
cycleReasoningEffortunit tests