[FEATURE] Option to reveal all thinking.
Preflight Checklist
- [x] I have searched existing requests and this feature hasn't been requested yet
- [x] This is a single feature request (not multiple features)
Problem Statement
When in thinking mode, Claude code hides thinking by default. You must hit <ctrl>o to see the thinking, but that locks the scrolling and you cant see how Claude is progressing.
In my use cases, Claude can go off the rails pretty easily for some code changes. Being able so see the thinking allows me to intervene more quickly and point it down a better thinking path to update the code.
Proposed Solution
It seems to me that in a earlier version of Claude Code, the thinking always showed up in the scrolling text.
Now it's hidden behind <ctrl>o. Can you create toggle key or option that allows the thinking to appear either 1) in the active scroll or 2) behind <ctrl>o so that the user can choose whether thinking is always visible in the scroll or not?
Alternative Solutions
I can't find a work around, other than toggling <ctrl>-o, but then I can't keep up with the scroll and claude still gets off track while I'm reading toggling and reading the past thinking....
Priority
High - Significant impact on productivity
Feature Category
Configuration and settings
Use Case Example
Example scenario:
1) I have asked Claude to implement a large scale refactor, perhaps broken into logical steps in a plan.
2) Claude "forgets" about parts of the implementation across a large code base
3) Claude starts thinking about how to implement the changes and starts toward a refactoring that would break other parts of the code.
When the thinking was displayed in the scrolling list, I could hit escape, remind Claude about the other parts of the code base that needed to be considered in designing the implementation
Now that the thinking is hidden by default, I can't keep up. When it was scrolling I could scan and prioritize what I read carefully. Now I have to hit ctrl-<o> for every "thought" to see if it's something I need to intervene on. Often, just by reviewing code changes, I approved several before I realize that Claude is starting to break the code with line changes. then I have to ask Claude to undo the changes and set Claude straight before proceeding again.
Additional Context
_No response_
11 Comments
Found 3 possible duplicate issues:
This issue will be automatically closed as a duplicate in 3 days.
🤖 Generated with Claude Code
+1, check my ccthink tool as a temporary solution and vote other issues, lets hope Anthropic recognizes it was a really bad move like they did with Todo list
+1
I ended up using https://github.com/aleks-apostle/claude-code-thinking-patch as a workaround (thanks to the link via @normalnormie). This is obviously a non-ideal solution, but it solves it in the short term.
Another vote for this, such a backwards move
This issue has been inactive for 30 days. If the issue is still occurring, please comment to let us know. Otherwise, this issue will be automatically closed in 30 days for housekeeping purposes.
bump
Looks like they've added a native fix for this...
/config > Verbose Output > True
This issue has been automatically closed due to 60 days of inactivity. If you're still experiencing this issue, please open a new issue with updated information.
This issue was closed despite human comments being present. This autoclose bot problem is reported in https://github.com/anthropics/claude-code/issues/16497
This issue has been automatically locked since it was closed and has not had any activity for 7 days. If you're experiencing a similar issue, please file a new issue and reference this one if it's relevant.