codex - 💡(How to fix) Fix No Persistent Auto-Compaction and Token Usage Status

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
RAW_BUFFERClick to expand / collapse

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

26.519.41501

What subscription do you have?

Pro

What platform is your computer?

Microsoft Windows NT 10.0.26200.0 x64

What issue are you seeing?

Auto-Compaction is not preserved inside the chat thread so I can see if it happened during work. It only shows up during the work and then goes away after work is completed. That is a real problem and needs to be retained in chat thread and not removed after the work is complete. It provides significant information for the users. Also, the auto-compaction method or resulting impact of it is brutal if not prepared. It seems to compact and flatten everything in the thread instead of prioritizing the most recent parts of the thread or existing work it's doing if it happens during the work. You would think it wouldn't compact any context window that the actual work being performed depends on to finish, but it does not. It compacts all of it and then mid-work completely messes up the remaining execution of the task. That is a major design flaw.

And in addition, and relevant, is the missing Token Usage Monitor that was at the bottom of the input next to the Model and Intelligence Dropdown. That was incredibly important and valuable and allowed the users to know when auto-compaction was about to occur, prep the ai workstream for it, and also get estimations on token usage during tasks to improve system optimizations.

What steps can reproduce the bug?

Just use the app in the same chat thread until it auto-compacts.

What is the expected behavior?

It should notify me of auto-compaction and then keep the auto-compaction part in the thread so we can see exactly when it happened. The token usage monitor should be at the bottom like it was before.

Additional information

I sure hope the removal of the monitor for token usage was not to prevent users from knowing their token burn. But after all the updates, I cannot understand any reason in favor of the Users/Customers to have removed that very useful token monitoring. This is very frustrating, and honestly, might be a deal breaker for me if it isn't fixed. Those auto-compactions are brutal at times on the amount of drift that can occur if not prepared when it happens.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING