[BUG] Token usage spikes even on off-hours
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?
I noticed that today (15:13:02 UTC Sunday, March 29, 2026) I (Max Plan) have experienced _massive_ spikes in token usage. 5 hour session is running out with a single cli session in 30 minutes of usage.
I have confirmed all other claude processes are stopped and others logged out completely. Each ~1m request is eating up 6-8% of my limit. This has never happened before.
I understand that the weekly prime hour limits are changing (even so, if it changes to this extent, that would be extremely surprising). I would not have expected changes on off-hours.
What Should Happen?
Behavior to remain the same prior to March 29.
Error Messages/Logs
Steps to Reproduce
Any claude usage.
Claude Model
Opus
Is this a regression?
Yes, this worked in a previous version
Last Working Version
_No response_
Claude Code Version
2.1.87
Platform
Anthropic API
Operating System
macOS
Terminal/Shell
Terminal.app (macOS)
Additional Information
Previous similar issues discuss the bug that was introduced and reportedly fixed last week, so I am assuming this behavior is unrelated.
This issue has 3 comments on GitHub. Read the full discussion on GitHub ↗