Where does the bug appear (feature/product)?
Cursor IDE
Describe the Bug
One of the latest updates changed the agent’s behavior. Previously, it would create a plan as a to-do list and immediately begin working on it, stopping only in the event of an emergency or when the outlined task was fully completed.
Now, the agent pauses and waits for something from the user after every sneeze. It created a plan as a to-do list and immediately stopped. I had a hard time convincing it to follow its own to-do list—it completed one item and stopped.
Moreover, the user is not informed in any way that any remaining to-do items remain, or that the work is in fact unfinished. I have to check this myself by scrolling through the chat.
Steps to Reproduce
n/a
Expected Behavior
As before: if the user specifically requested approval of the action plan, then after preparing the to-do list, it is appropriate to request their consent/corrections. If they haven’t requested it, then after creating the to-do list, it immediately begins step-by-step execution without unnecessary pauses. Only in cases of unclear situations or possible work loops.
Operating System
MacOS
Current Cursor Version (Menu → About Cursor → Copy)
Version: 2.0.69 (Universal)
VSCode Version: 1.99.3
Commit: 63fcac100bd5d5749f2a98aa47d65f6eca61db30
Date: 2025-11-07T18:21:29.650Z
Electron: 37.7.0
Chromium: 138.0.7204.251
Node.js: 22.20.0
V8: 13.8.258.32-electron.0
OS: Darwin arm64 24.6.0
For AI issues: which model did you use?
Auto
For AI issues: add Request ID with privacy disabled
49f836f2-296f-44b1-932c-1a5ac65238cf
Does this stop you from using Cursor
Sometimes - I can sometimes use Cursor