### Preflight Checklist - [x] I have searched [existing requests](https://github.com/anthropics/claude-code/issues?q=is%3Aissue%20label%3Aenhancement) and this feature hasn't been requested yet - [x] This is a single feature request (not multiple features) ### Problem Statement Recently as a part of a new update, the expanded thinking block in the desktop app shows only extremely vague one-line summaries like, here is an example from a chat I had about asking claude to find the checkmate on a

[Desktop App] Thinking blocks show only vague 1-line summaries instead of detailed reasoning

Status Open

Maintainer reply None cached

Activity 0 comments · opened Sep 6, 2026

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

Recently as a part of a new update, the expanded thinking block in the desktop app shows only extremely vague one-line summaries like, here is an example from a chat I had about asking claude to find the checkmate on a chess board:

"Working out the best move in a chess position."
"Testing knight moves for a forced checkmate pattern."
"Rechecking piece placement and probing queen sacrifice lines"

Previously, the thinking block showed the actual detailed reasoning process -- the step-by-step logic, considerations, and self-corrections Claude was working through in real time.

Why this impacts me:

The detailed thinking display served a real functional purpose: catching errors early meant saving token usage. When I could see Claude's actual reasoning, I could catch a wrong assumption or flawed approach mid-thinking and interrupt before the full (incorrect) output was generated. With vague summaries, there is no signal that the reasoning is going in the wrong direction until the full response is already produced and usage is spent.

What I've tried:

Added "showThinkingSummaries": true and "verbose": true to ~/.claude/settings.json -- no effect on the desktop app (confirmed these only work in Claude Code CLI)
The thinking block IS expanded -- the content inside it is just vague

Environment: macOS desktop app

Related: #52376 (closed, CLI-focused)

Proposed Solution

Either:

Restore full detailed reasoning in the thinking block (previous behavior before the update), or
Add a toggle in desktop app Settings to enable full thinking display

The showThinkingSummaries and verbose settings in ~/.claude/settings.json should also apply to the desktop app, not only to Claude Code CLI.

Alternative Solutions

_No response_

Priority

High - Significant impact on productivity

Feature Category

Configuration and settings

Use Case Example