Codex doesn’t think proactively enough and usually requires explicit instructions from me to continue working.

Resolved 💬 3 comments Opened May 12, 2026 by CloudSen Closed May 12, 2026

💡 Likely answer: A maintainer (github-actions[bot], contributor) responded on this thread — see the highlighted reply below.

What version of the Codex App are you using (From “About Codex” dialog)?

26.506.31421 (2620)

What subscription do you have?

GPT PRO 100$/month

What platform is your computer?

Darwin 25.4.0 arm64 arm

What issue are you seeing?

Codex doesn’t think proactively enough and usually requires explicit instructions from me to continue working.
After my instructions, Codex always responds with things like “Yes, agreed,” “Correct,” or “Semantically yes.”
It still doesn’t think any further.

What steps can reproduce the bug?

Uploaded thread: 56419099-8fd7-45c1-87a4-5b48e178197d

What is the expected behavior?

_No response_

Additional information

_No response_

View original on GitHub ↗

3 Comments

github-actions[bot] contributor · 3 months ago

Potential duplicates detected. Please review them and close your issue if it is a duplicate.

#20958
#21410
#22219

Powered by Codex Action

etraut-openai contributor · 3 months ago

Thanks for the bug report. This appears to be model behavior, not a Codex harness issue. We'll share the feedback with the team responsible for training our codex models. We're tracking this feedback and similar reports so we can improve future models. And thanks for using the /feedback command. That additional information is useful.

etraut-openai contributor · 3 months ago

You may be able to steer the model toward your desired behavior through instructions in your AGENTS.md file.