Summary

Agent failed before reply: All models failed (1): openai-codex/gpt-5.2-codex: Provider openai-codex is in cooldown (all profiles unavailable) (rate_limit).
Logs: openclaw logs --follow

Agent failed before reply: HTTP 401: You have insufficient permissions for this operation. Missing scopes: model.request. Check that you have the correct role in your organization (Reader, Writer, Owner) and project (Member, Owner), and if you're using a restricted API key, that it has the necessary scopes.
Logs: openclaw logs --follow

I have 100% usage available. My agent has been having issues the last 24 hours so I did a total clean reinstall. It was working fine, but when moving some of the old files back to the folders for memory and work tasks, this issue appears again.

I'm getting circular responses with Claude Code and Gemini to try and fix it. There must be something else that's being misrepresented within the error messages (I have re-authed codex, restarted gateway etc dozens of times at this point).

Steps to reproduce

1. Login via Codex
2. Try to use 5.2 or 5.3-codex
3. Fail
4. Fallback to a different model

Expected behavior

The OpenAI OAuth should work, and it was working prior to 20th February fine, since then something does not seem to like me.

Actual behavior

Every time I try to use an openai-codex/ model I get a 401 error despite what my actual usage allowance is.

OpenClaw version

2026.2.21

Operating system

macOS Tahoe

Install method

npm global

Logs, screenshots, and evidence

Impact and severity

Severity: High

Additional information

This started after version 2026.2.19