Cursor's queued message doesn't work well with Codex models

Ideas Feature Requests
,
1

Where does the bug appear (feature/product)?

Cursor IDE

Describe the Bug

Codex models are marketed to support interruption and steering messages from users but seems like Cursor doesn’t support this very well. Queued messages only get sent when Codex is done with its tasks.

Steps to Reproduce

Queuing messages when using Codex models.

Expected Behavior

Queued messages to be sent after some tool calls

Operating System

MacOS

Version Information

Version: 3.1.15 (Universal)
VSCode Version: 1.105.1
Commit: 3a67af7b780e0bfc8d32aefa96b8ff1cb8817f80
Date: 2026-04-15T01:46:06.515Z (5 days ago)
Layout: editor
Build Type: Stable
Release Track: Default
Electron: 39.8.1
Chromium: 142.0.7444.265
Node.js: 22.22.1
V8: 14.2.231.22-electron.0
OS: Darwin arm64 25.3.0

For AI issues: which model did you use?

Codex 5.3 Extra High

Does this stop you from using Cursor

Sometimes - I can sometimes use Cursor

6

Hey! Thanks for the detailed report. What you’re describing is actually the current expected behavior for queued messages: they’re held until the agent’s current turn finishes, then dispatched automatically. This applies to all models, including Codex.

The closest existing mechanism for what you want is Cmd+Enter (or Ctrl+Enter on Windows/Linux), which sends your message immediately and the agent considers it right away alongside its current work. You can read more about it in the Agent docs under “Immediate messaging”.

That said, what you’re describing (the agent picking up queued messages mid-turn, between tool calls, without stopping) is a step beyond what the current queue or immediate-send provides. A few others have requested similar behavior: