Show model and reasoning effort details for Codex subagents

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

Version 26.901.51231

What subscription do you have?

Yes

What platform is your computer?

Darwin 25.5.0 arm64 arm

What issue are you seeing?

Codex currently provides limited visibility into subagent configuration and execution. Users cannot easily determine which model a subagent used or what reasoning-effort setting was applied.

This makes it difficult to understand differing results, troubleshoot unexpected behavior, and compare subagent runs.

What steps can reproduce the bug?

Feedback ID: 01a08747-fa5c-72c0-b6f9-7ac903c81fcc

What is the expected behavior?

Show an expandable details panel for each subagent, including where available:

Model used

Reasoning-effort setting

Subagent role or instructions

Tools and permissions

Execution time

Token usage

This information would improve debugging, reproducibility, and configuration decisions without cluttering the main interface.

Additional information

A compact panel could appear both while a subagent is running and in completed task history. Sensitive prompts or internal information could be summarized or omitted as appropriate.

Start by tracing the subagent execution view and completed task history described in the issue, then identify where subagent configuration and run details are represented. Done means each subagent can expose an expandable panel with available model, reasoning effort, role, permissions, execution time, and token usage without cluttering the main interface.