3 free ultrareview runs consumed but crashed due to server-side rate limit — no findings returned

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?

All 3 of my free ultrareview runs (expiring May 5, 2026) crashed without returning any findings. The error was a server-side rate limit, not caused by my code or usage.

Each run launched successfully ("Free ultrareview 1/2/3 of 3") but all returned:
"Review crashed before producing findings. See session logs for details."

What Should Happen?

Either the reviews should complete successfully, or runs that fail due to server-side errors should not count against the free allocation.

Error Messages/Logs

API Error: Server is temporarily limiting requests (not your usage limit) · Rate limited

  {"error":"Review crashed before producing findings. See session logs for details."}

Steps to Reproduce

. Create a branch with code changes (e.g. git checkout -b review/ai-agent <old-commit> then add files from main)

1. Run /ultrareview — launches as "Free ultrareview 1 of 3"
2. Switch to second branch, run /ultrareview — "Free ultrareview 2 of 3"
3. Switch to third branch, run /ultrareview — "Free ultrareview 3 of 3"
4. All 3 return the rate limit error within minutes
5. No findings produced, but all 3 free credits are consumed

Claude Model

Opus

Is this a regression?

I don't know

Last Working Version

_No response_

Claude Code Version

2.1.126 (Claude Code)

Platform

Anthropic API

Operating System

Windows

Terminal/Shell

PowerShell

Additional Information

Session IDs for the 3 failed runs:

• Run 1 (AI agent, 46 files, 11012 lines): session_017g374xjuBTzbMKm9gZ3cSA
• Run 2 (Financial, 14 files, 3911 lines): session_01T5zhh42JmojrStf6vSYkqE
• Run 3 (Email, 22 files, 6636 lines): session_01WrEoZPQVLkZXMNBuomycnH

Request: Could the 3 free ultrareview credits be restored since the failure was entirely server-side?