[Bug] Anthropic API Error: Opus 5 safeguards flagging all messages with reasoning_extraction error

**Bug Description**
No matter what I do or say, everything I try to do with Opus (Claude Code cli on Ubuntu Linux), I get the following error message.

API Error: Opus 5'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 Opus 5.

Double press esc to edit your last message, or try a different model with /model.

Send feedback with /feedback or learn more: https://support.claude.com/en/articles/16049681

Details: `[reasoning_extraction]`

Request ID: req_011CfAAM3s9t5Djx1PmMdzbE

I have tried compacting. It just started happening today, and happens across every session I have open. It can't be because of one particular session.

**Environment Info**
- Platform: linux
- Terminal: vte-based
- Version: 2.1.275
- Feedback ID: 5e4dc5aa-e90d-4d8c-9017-7726fa9c19ae

**Errors**
```json
[]
```

No source file or test is named. Start by reproducing the reasoning_extraction error on Ubuntu Linux with Claude Code 2.1.275 and Opus 5, then trace how the CLI handles the API response; done means identifying whether the failure is in Claude Code or the upstream API and verifying the resulting fix or dependency.

Written by the indexing model from the issue text.