CLI `/goal` does not stop after explicit user instruction to wrap up and stop

What version of Codex CLI is running?

0.130.0

What subscription do you have?

GPT Pro

Which model were you using?

gpt-5.5-codex

What platform is your computer?

Linux

What terminal emulator and version are you using (if applicable)?

CLI

What issue are you seeing?

When using the Codex CLI /goal feature, the agent appears unable to stop cleanly even after I explicitly instruct it to wrap up the current progress and stop.

Instead of terminating the goal/task loop, the CLI repeatedly prints:

• Stopped.

What steps can reproduce the bug?

Start with an objective using /goal, and wait enough time (before codex thinks itself achieved the goal), send queued message and ask it to wrap up and stop.

What is the expected behavior?

When the user explicitly asks the agent to stop, for example:

Please wrap up the current progress and stop.

the Codex CLI should:

summarize the current progress,
stop the active goal loop cleanly,
return control to the user,
not repeatedly emit Stopped.,
not require an external interruption.

Additional information

<img width="2047" height="1189" alt="Image" src="https://github.com/user-attachments/assets/7704ebf2-ca06-4321-9458-ec8c1a56eb87" />