[BUG] ultrathink-triggered activation of high effortLevel not accurately reflected in UI

Preflight Checklist

• [x] I have searched existing issues and this hasn't been reported yet
• [x] This is a single bug report (please file separate reports for different bugs)
• [x] I am using the latest version of Claude Code

What's Wrong?

The UI has a status message beside the spinner verb that shows (thinking with <effortLevel> effort).
There is also a thinking level indicator on the prompt input field top border (▪▪▪).

When ultrathink is in the prompt

After the ultrathink keyword is introduced in a prompt, there is an ephemeral activation update on the right side of status line that says "Effort set to high for this turn". But, the thinking level indicator still shows the default effortLevel.

<img width="1561" height="310" alt="Image" src="https://github.com/user-attachments/assets/8f472cc9-48ec-4062-84a4-c31b1824b8c2" />

When the prompt is submitted

After submission of the prompt, when the model is thinking, the status message still shows the currently default effortLevel.

<img width="1561" height="310" alt="Image" src="https://github.com/user-attachments/assets/20734c27-86b0-4f0e-9d40-2d2024d79d1f" />

What Should Happen?

When ultrathink is in the prompt

The thinking level indicator in the UI should indicate high (3 activated dots).

When the prompt is submitted

The status message beside the spinner verb should show (thinking with high effort)

Error Messages/Logs

Steps to Reproduce

Prepare any prompt message with the ultrathink activation word in it and submit it.

Track the thinking level indicator and the thinking status message beside the spinner verb, when it occurs.

Claude Model

None

Is this a regression?

I don't know

Last Working Version

_No response_

Claude Code Version

2.1.71

Platform

Anthropic API

Operating System

macOS

Terminal/Shell

Other

Additional Information

_No response_