Codex requires constant attention

What version of Codex is running?

codex-cli 0.66.0

What subscription do you have?

Plus

Which model were you using?

gpt-5.1-codex-max high

What platform is your computer?

Linux 6.17.0-7-generic x86_64 unknown

What issue are you seeing?

It's very frustrating that codex requires constant attention, rarely completes tasks before coming back with the same darn summary + next steps, and requires me to feed it every 1-5 minutes with generic 'go ahead', 'please continue', 'please implement X' messages.

It frequently mentions 'time constraints' or 'scope constraints', and since these did not come from me I can only assume they're from OpenAI. Please lighten up on the constraints, this behaviour is killing my productivity.

What steps can reproduce the bug?

Uploaded thread: 019afb01-ab7e-7282-9396-

What is the expected behavior?

Complete the assigned task before coming back and asking for user input.

Additional information

_No response_