[BUG] Assistant suggests/uses paid "extra usage" after being told not to, repeatedly

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 the 5-hour plan usage limit is reached, Claude Code describes pay-as-you-go "extra usage" (overage) as available and something the user could draw from to keep working — even after I've explicitly told it in a prior conversation not to do this once the free limit is hit. This has now happened more than once across sessions, suggesting it's not just a one-off phrasing slip but a recurring pattern — possibly because there's no persistent, product-level setting to suppress this suggestion, so it falls entirely on conversational memory to avoid repeating it. Expected: once a free limit is reached, the assistant should state the limit/reset time plainly without framing paid overage as the solution, and ideally there'd be an account-level setting to disable even mentioning it.

What Should Happen?

It should stop and tell me that limit has reached and stopp using paid credits.

Error Messages/Logs

Steps to Reproduce

Nothing to pate here its on Claudes end.

Claude Model

None

Is this a regression?

Yes, this worked in a previous version

Last Working Version

_No response_

Claude Code Version

NA

Platform

Anthropic API

Operating System

macOS

Terminal/Shell

Terminal.app (macOS)

Additional Information

_No response_