Documentation Type
Incorrect/outdated documentation
Documentation Location
https://code.claude.com/docs/en/interactive-mode
Section/Topic
Keyboard shortcuts table, Ctrl+L row
Current Documentation
The interactive mode docs currently describe Ctrl+L as clearing typed input:
Ctrl+L | Clear prompt input and redraw screen | Clears typed text and forces a full terminal redraw. Conversation history is kept. Use this to recover if the display becomes garbled or partially blank
The keybindings reference also maps Ctrl+L to chat:clearInput:
chat:clearInput | Ctrl+L | Clear prompt input and force a full screen redraw. In fullscreen rendering, press twice within two seconds to run /clear
The fullscreen page repeats the old first-press behavior:
Press Ctrl+L twice within two seconds to run /clear and start a new conversation. The first press clears the input box and shows a hint; the second press clears the conversation.
However, the v2.1.126 changelog says the behavior changed:
Fixed Ctrl+L clearing the prompt input - it now only forces a screen redraw, matching readline behavior
What's Wrong or Missing?
The Claude Code docs still document the pre-v2.1.126 behavior where pressing Ctrl+L clears prompt input. In v2.1.126, Ctrl+L was fixed to only force a screen redraw and no longer clear the prompt input.
This leaves the keyboard shortcut documentation internally inconsistent with the changelog and may cause users to expect prompt text loss when using the redraw shortcut.
Suggested Improvement
Update the affected Ctrl+L descriptions to match the v2.1.126 behavior.
For example, change the interactive mode row from:
Clear prompt input and redraw screen
to:
Redraw screen
And change the description from:
Clears typed text and forces a full terminal redraw.
to:
Forces a full terminal redraw without clearing typed text.
Also update the keybindings reference so Ctrl+L is no longer documented as chat:clearInput unless that action name remains correct internally for compatibility reasons. If the fullscreen double-press behavior still exists, clarify that the first press redraws the screen and shows the /clear hint without clearing the input box.
Impact
Medium - Makes feature difficult to understand
Additional Context
Affected Pages:
Total scope: 3 documentation pages appear outdated, with the changelog providing the current behavior reference.
Documentation Type
Incorrect/outdated documentation
Documentation Location
https://code.claude.com/docs/en/interactive-mode
Section/Topic
Keyboard shortcuts table,
Ctrl+LrowCurrent Documentation
The interactive mode docs currently describe
Ctrl+Las clearing typed input:The keybindings reference also maps
Ctrl+Ltochat:clearInput:The fullscreen page repeats the old first-press behavior:
However, the v2.1.126 changelog says the behavior changed:
What's Wrong or Missing?
The Claude Code docs still document the pre-v2.1.126 behavior where pressing
Ctrl+Lclears prompt input. In v2.1.126,Ctrl+Lwas fixed to only force a screen redraw and no longer clear the prompt input.This leaves the keyboard shortcut documentation internally inconsistent with the changelog and may cause users to expect prompt text loss when using the redraw shortcut.
Suggested Improvement
Update the affected
Ctrl+Ldescriptions to match the v2.1.126 behavior.For example, change the interactive mode row from:
to:
And change the description from:
to:
Also update the keybindings reference so
Ctrl+Lis no longer documented aschat:clearInputunless that action name remains correct internally for compatibility reasons. If the fullscreen double-press behavior still exists, clarify that the first press redraws the screen and shows the/clearhint without clearing the input box.Impact
Medium - Makes feature difficult to understand
Additional Context
Affected Pages:
Ctrl+Lrow says it clears prompt input and typed textchat:clearInputrow mapsCtrl+Lto clearing prompt input and forcing redraw/cleardocs say the firstCtrl+Lpress clears the input boxCtrl+Lnow only forces a screen redrawTotal scope: 3 documentation pages appear outdated, with the changelog providing the current behavior reference.