[BUG] Serious usage tracking bug: quota increased from 64% to 100% with no active tasks
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 found a serious bug in the usage tracking system, and it has significantly affected my user experience.
Today, I only opened two conversations. In each conversation, I fixed one simple issue. After both tasks were completed, my 5-hour usage limit had reached 64%.
At that point, all conversation tasks had already ended, and there were no background tasks running. However, after a little more than one hour, my 5-hour usage suddenly reached 100%, even though there was still about 1.5 hours left before the 5-hour reset.
I am 100% sure that during this period, my account was not used by anyone else, and I was not logged in or using it on any other computer or terminal.
But somehow, around 30% of my usage disappeared for no reason.
Please investigate this issue, as it seems like a serious usage/quota tracking bug.
What Should Happen?
Thinking back, I feel like my recent usage consumption has been abnormal. My 5-hour quota has been getting used up very quickly, and my weekly quota often reaches around 45% by the second day, even though I have not been using it heavily.
This never used to happen before, and it only started occurring recently after the official usage limits were loosened/expanded.
Error Messages/Logs
Steps to Reproduce
I was only using it normally — opening conversations, asking questions, and making modifications. I only opened two conversations in total, and each conversation only involved about 1–3 exchanges. That was all.
Claude Model
None
Is this a regression?
Yes, this worked in a previous version
Last Working Version
_No response_
Claude Code Version
v2.1.142
Platform
Anthropic API
Operating System
macOS
Terminal/Shell
Terminal.app (macOS)
Additional Information
_No response_
This issue has 2 comments on GitHub. Read the full discussion on GitHub ↗