[Bug] Anthropic API Error: Content filtered by safety policy (reasoning_extraction)

**Bug Description**
https://xxxx.atlassian.net/browse/DATA-1258 Can we take a look at this?

⏺ 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.

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/15363606

Details: `[reasoning_extraction]`

Request ID: req_011Cf98wHsj1me61Xkx9N5k3

**Environment Info**
- Platform: darwin
- Terminal: iTerm.app
- Version: 2.1.273
- Feedback ID: b96578ef-67d6-4eee-b08d-1d5a7802b5f1

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

Start with the linked DATA-1258 report and the provided feedback ID and request ID, then investigate the reported reasoning_extraction response on darwin in Claude Code. Done should be defined as identifying why safe conversations are rejected and documenting or verifying a reproducible fix for the affected API behavior.

Written by the indexing model from the issue text.