[BUG] Korean IME input causes character decomposition after language switching

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?

When switching between Korean and English input using the language toggle key and then typing Korean characters, the characters appear decomposed into separate consonants and vowels instead of being properly composed. For example, typing "한" may appear as "ㅎㅏㄴ" instead of properly formed characters.

What Should Happen?

After toggling between Korean and English input methods, Korean characters should be properly composed and displayed as complete syllables. The IME should correctly handle the composition events, combining consonants and vowels into complete Hangul syllables.

Error Messages/Logs

Steps to Reproduce

1. Launch Claude Code in iTerm2 terminal
2. Type an English word in the command input field
3. Toggle input language to Korean (using input method switcher)
4. Type Korean characters
5. Observe that the first syllable's consonants and vowels are separated

• Expected: Complete syllables like "한글"
• Actual: First syllable decomposed like "ㅎㅏㄴ글"

Claude Model

Sonnet (default)

Is this a regression?

Yes, this worked in a previous version

Last Working Version

_No response_

Claude Code Version

2.0.55

Platform

Anthropic API

Operating System

macOS

Terminal/Shell

iTerm2

Additional Information

• iTerm2: 3.6.6
• macOS: Tahoe 26.1

https://github.com/user-attachments/assets/c12ac799-7079-4fc3-9d10-6603465c05c8