Auto context compaction keeps running indefinitely

Open 💬 2 comments Opened Apr 24, 2026 by 15230745073
💡 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)?

26.423.10653

What subscription do you have?

Business

What platform is your computer?

Microsoft Windows NT 10.0.26200.0 x64

What issue are you seeing?

Codex Windows app version: 26.423.10653
OS: win11 25H2 26200.8246
session id: 019dbe30-9e5f-7302-8b84-c8d70550fc1e
Workspace / plan: Business
Agent mode: Windows native
Project path: Windows filesystem
Symptom:

  • Auto context compaction keeps running indefinitely
  • Stop does not clear state
  • Sending another message re-triggers compaction
  • Background/context panel does not visibly shrink
  • 5-hour usage dropped from 76% to 39%

Steps tried:

  • Stopped generation manually
  • Restarted app
  • Removed / moved affected session file
  • Compact by manual

What steps can reproduce the bug?

Uploaded thread: 019dbe30-9e5f-7302-8b84-c8d70550fc1e

What is the expected behavior?

_No response_

Additional information

_No response_

View original on GitHub ↗

2 Comments

github-actions[bot] contributor · 4 months ago

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

  • #18363
  • #18827
  • #18568

Powered by Codex Action

mahdikayvan · 3 months ago

@15230745073 Hey — this issue felt painfully familiar. It is very close to the trust/drift problem I’m trying to solve.

I’m building AgaveCore, and I’d like to let a few early people use it for free while I shape it into something stronger.

It helps when a coding agent starts to drift, loses important context, or gives you an output you do not fully trust.

In practice, it works through two habits:

  • tell the agent to remember-progress or remember-this when something important should not be lost
  • use consult-brain when the current task needs more grounded feedback on what to do next

Over time, as useful journals, decisions, and failures pile up, the product should help the agent make sharper decisions and take better actions with less drift.

If this sounds relevant, reply here and I’ll send the exact steps. You can also use the email or Telegram on my profile.

Cheers,
Mads.