[BUG] Incorrect rendering of user and assistant messages (user message overlaps with assistant message)

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 I use Ctrl+G to edit a User Message in an external editor (e.g VS Code), the message is returned to Claude Code in such a way that the message overlaps with the assistant response in such a way as to make both message partially unreadable
• It looks like additional line breaks are being inserted/not rendered correctly
• This usually requires at least one assistant message to be in the conversation to trigger.

What Should Happen?

• User message should display exactly as written, without overlapping assistant message

Error Messages/Logs

• Image attached

<img width="1067" height="720" alt="Image" src="https://github.com/user-attachments/assets/88e1c844-c59c-46e3-b33a-e140df8c02d2" />

Steps to Reproduce

1. Start a conversation with Claude
2. Ensure that a few messages have been exchanged
3. Edit a user message in an external editor (e.g. VS Code)
4. Close external editor - message appears in terminal, correctly formatted
5. Press enter - user message now has additional line breaks, and overlaps with assistant message

Claude Model

Opus

Is this a regression?

Yes, this worked in a previous version

Last Working Version

_No response_

Claude Code Version

2.1.92

Platform

Anthropic API

Operating System

Windows

Terminal/Shell

PowerShell

Additional Information

• Replicable in all terminals on Windows (Powershell, Bash)
• This has been broken for around 2-3 weeks - I don't have the exact version number when the regression happened