Describe the Bug
When the agent runs terminal commands, it sometimes thinks there was no output when in reality the command hadn’t finished. So then it tries again, and reads the previous terminal output and then sees the output. But the command is running again. So this causes issues and uses a couple more tool calls than necessary.
Steps to Reproduce
It runs RSpec commands, and sometimes it’ll still be running and then the agent will write:
The last command gave no output. I’ll rerun it with verbose RSpec formatting to see what’s happening with the tests.
Expected Behavior
Runs only once and waits for command to finish.
Operating System
MacOS
Current Cursor Version (Menu → About Cursor → Copy)
Version: 1.1.3
VSCode Version: 1.96.2
Commit: 979ba33804ac150108481c14e0b5cb970bda3260
Date: 2025-06-15T06:35:49.230Z
Electron: 34.5.1
Chromium: 132.0.6834.210
Node.js: 20.19.0
V8: 13.2.152.41-electron.0
OS: Darwin arm64 24.2.0
Additional Information
It’s been happening to me since around 0.50?
I’ve tried closing terminals, etc. in case it’s some terminal state thing.
Maybe it happens more when scrolling up to see terminal output, or on the first terminal command of new chats?
Does this stop you from using Cursor
No - Cursor works, but with this issue