Unable to cleanup agents (limit reached)

[kirso]

What issue are you seeing?

I hit the sub-agent limit, so I’m running validation/tests directly in this shell to keep momentum. I’ll report exact failures if anything breaks.

What steps can reproduce the bug?

Exhaust all agents in the thread, try to spawn more

What is the expected behavior?

Ability to cleanup agents, persist IDs between compactions

Additional information

No response

[github-actions]

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

• Codex startup sub-agent will restart a bunch of mcp, but it will not be automatically cleared after closing the sub-agent. #12976
• High invisible output churn when idle (request for regression test or to fix the suppressed bug) #11857
• Add automatic lifecycle cleanup for completed/idle sub-agents (and unused MCP clients) #12335

Powered by Codex Action

[jif-oai]

Hi, on which version are you? A fix was merged yesterday for this and will be in the next release