[BUG] [BUG] False positive safeguard trigger on safe coding conversation (reasoning_extraction)

Preflight Checklist

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

What's Wrong?

Normal coding session (building pixel art project), no sensitive content. Asked about progress and got flagged. Error spreads to every new conversation window. Both Opus 5 and Fable 5.1 affected.

What Should Happen?

Normal coding conversations should not trigger safeguards.

Error Messages/Logs

API Error: Opus 5's safeguards flagged this message. Details: [reasoning_extraction]

 API Error: Fable 5.1's safeguards flagged this message (https://www.anthropic.com/legal/aup). This sometimes happens with safe, normal conversations. Claude Code can't respond to this message with Fable 5.1.

Steps to Reproduce

1. Have custom user preferences set (in Chinese) 2. Work on a normal coding project 3. Ask about progress 4. Get blocked on all subsequent conversations

Claude Model

None

Is this a regression?

Yes, this worked in a previous version

Last Working Version

No response

Claude Code Version

globalVersion: 2.1.276

Platform

Anthropic API

Operating System

Windows

Terminal/Shell

Windows Terminal

Additional Information

No response

The report names no repository file, test, or code entry point. Start by reproducing the safeguard trigger with Claude Code 2.1.276 on Windows using the described preferences and coding conversation, then compare behavior with the previous version if available. Done means safe coding conversations no longer trigger the safeguard or block subsequent windows.