[BUG]

Resolved 💬 3 comments Opened Apr 30, 2026 by PAK43 Closed May 3, 2026

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 usage limit exceeded, Silent Session Termination — No Warning, No Error Message

When a session time or usage limit is reached, Claude frequently stops responding with no message or indication whatsoever. The UI remains in a "processing" state for 10–15+ minutes, leading me to believe the model is still working — when in reality the session has already been silently terminated.

OBSERVED BEHAVIOR

  • No warning as the session approaches its limit
  • No error or termination message at cutoff (occasionally a message appears, but this is inconsistent)
  • UI stays in a "thinking" state indefinitely after cutoff

This issue appears to have worsened in the last 2 days, suggesting a possible regression or recent change to limit-handling logic.

Silent failure without user feedback is a critical UX flaw.

What Should Happen?

EXPECTED BEHAVIOR
At minimum: a clear, deterministic message when the session limit is reached.

Ideally:

  • Progressive warnings as usage approaches the limit (e.g., at 75%, 90%, 95%)
  • A visible indicator of remaining session time or quota
  • Graceful termination — allow the current response to complete or provide a short grace window
  • Option to resume or recover context after cutoff

Error Messages/Logs

No messages. That is the complaint.

Steps to Reproduce

Exceed usage limits in claude code and it just stops untill usage time resets.

Claude Model

Sonnet (default)

Is this a regression?

I don't know

Last Working Version

_No response_

Claude Code Version

maybe 2.1.119

Platform

Anthropic API

Operating System

Windows

Terminal/Shell

Other

Additional Information

This is the claude app, I have no idea what api or shell. it is a native windows program, that is all I know about it. Much more documtation would be appreciated also.

View original on GitHub ↗

This issue has 3 comments on GitHub. Read the full discussion on GitHub ↗