Where does the bug appear (feature/product)?
Cursor IDE
Describe the Bug
The recent changes to how work is nested in the chat have made it considerably harder to review and trace changes.
1. The new nested work view is difficult to navigate
Previously, it was easy to scroll through the chat and see which files were changed, in what order, and trace the work step by step.
With the new collapsed sections, such as “Worked for 15m,” reviewing previous work now requires finding the relevant section, expanding it, navigating through the nested content, and collapsing it again afterward.
The result feels like an increasingly deep tree of nested elements, which makes reviewing and understanding the sequence of changes much more difficult. In my opinion, this is a significant usability regression from the previous layout.
2. Changed files are no longer appearing reliably
There was already a known issue where clicking a changed file inside the chat would not automatically open it.
After the latest update, some of my recent sessions do not appear to show the changed files in the chat at all. This makes it difficult to inspect the agent’s work or quickly navigate to the affected files.
3. Subagent conversations behave inconsistently
I have also noticed that subagents sometimes behave like full, independent agents, allowing the user to continue chatting with them directly.
This could be a useful feature, but the behavior is inconsistent. In some cases, a message input is available and the conversation can continue. In other cases, the subagent output is read-only and no input field is shown.
It would be helpful to clarify whether continued interaction with subagents is an intended feature and, if so, make the behavior consistent across sessions.
Steps to Reproduce
.
Screenshots / Screen Recordings
Operating System
MacOS
Version Information
Version: 3.11.13 (Universal)
VS Code Extension API: 1.125.0
Commit: 3f21b08f0b436a07be29fbfe00b304fa15553350
Date: 2026-07-10T01:45:28.254Z
Layout: IDE
Build Type: Stable
Release Track: Default
Electron: 40.10.3
Chromium: 144.0.7559.236
Node.js: 24.15.0
V8: 14.4.258.32-electron.0
xterm.js: 6.1.0-beta.256
For AI issues: which model did you use?
CPT-5.6 Sol 1M Max
Does this stop you from using Cursor
Sometimes - I can sometimes use Cursor