Preflight Checklist
What's Wrong?
Description
Receiving rate limit errors on Claude Code even though usage counters show 0-1% consumed. The service is blocking requests despite available quota on a paid Max plan.
What Should Happen?
Errors observed
API Error: Rate limit reached
You've hit your limit · resets 1pm (America/Fortaleza)
Environment
- Plan: Max
- Model tested: Opus 4.6, Sonnet 4.6
- Tool: Claude Code
- OS: (please fill in)
- Claude Code version: (run
claude --version)
Usage counters at time of error (screenshots attached)
- Current session: 1% used
- Weekly limit (All models): 0% used
- Weekly limit (Sonnet only): 0% used
- Last updated: just now
Error Messages/Logs
Steps to Reproduce
Steps to reproduce
- Open Claude Code with Max plan
- Run a command using
/prd-creator skill (e.g. read /app/dashboard and generate its PRD)
- Error appears immediately, even with minimal scope commands
- Switching from Opus 4.6 to Sonnet 4.6 did not resolve
- Opening a new conversation did not resolve
- Reducing command scope (from full codebase to single folder) did not resolve
Expected behavior
Requests should be processed normally since usage counters show available quota.
Actual behavior
All requests are blocked with rate limit errors regardless of usage level shown in Settings → Usage.
Additional context
The issue persists across model switches (Opus → Sonnet) and new conversations. Usage dashboard confirms limits are not exhausted. This appears to be a mismatch between the rate limit enforcement logic and the actual usage counters displayed to the user.
Screenshots of usage counters showing 0% available attached.
Claude Model
Sonnet (default)
Is this a regression?
Yes, this worked in a previous version
Last Working Version
No response
Claude Code Version
Versão 1.1.7714 (3bd6f6)
Platform
Anthropic API
Operating System
Windows
Terminal/Shell
Windows Terminal
Additional Information
No response
Preflight Checklist
What's Wrong?
Description
Receiving rate limit errors on Claude Code even though usage counters show 0-1% consumed. The service is blocking requests despite available quota on a paid Max plan.
What Should Happen?
Errors observed
API Error: Rate limit reachedYou've hit your limit · resets 1pm (America/Fortaleza)Environment
claude --version)Usage counters at time of error (screenshots attached)
Error Messages/Logs
Steps to Reproduce
Steps to reproduce
/prd-creatorskill (e.g.read /app/dashboard and generate its PRD)Expected behavior
Requests should be processed normally since usage counters show available quota.
Actual behavior
All requests are blocked with rate limit errors regardless of usage level shown in Settings → Usage.
Additional context
The issue persists across model switches (Opus → Sonnet) and new conversations. Usage dashboard confirms limits are not exhausted. This appears to be a mismatch between the rate limit enforcement logic and the actual usage counters displayed to the user.
Screenshots of usage counters showing 0% available attached.
Claude Model
Sonnet (default)
Is this a regression?
Yes, this worked in a previous version
Last Working Version
No response
Claude Code Version
Versão 1.1.7714 (3bd6f6)
Platform
Anthropic API
Operating System
Windows
Terminal/Shell
Windows Terminal
Additional Information
No response