Rate-limit banner typo + opaque limits

What version of Codex is running?

codex-cli 0.19.0

Which model were you using?

gpt-5

What platform is your computer?

Darwin 24.5.0 arm64 arm

What steps can reproduce the bug?

1. Open codex CLI using an OpenAI Plus/Pro account.
2. Run normal workflows for ~1–2 hours (light coding/planning is enough) until rate limits are hit.
3. Observe the banner that appears when the cap is reached.

What is the expected behavior?

• Correct grammar in the banner.
• Transparent limits displayed at the moment of failure and queryable at any time:

Remaining vs. total requests/runs, tokens (if applicable), and per-model breakdown.
Time-to-reset for the rolling 5-hour window and the weekly window (absolute timestamps).

• Early warnings at ~80% and ~95% of the cap with actionable guidance (e.g., suggest lowering reasoning effort or switching models).
• A command (e.g., codex usage) that shows the same counters on demand.

What do you see instead?

• Banner text shows a typo/grammar error:

🖐 You've hit usage your usage limit. Limits reset every 5h and every week.

• No counters, no per-model breakdown, and no time-to-reset.
• No early warning before the hard stop.
• No obvious command to check current usage status.

Additional information

Users cannot plan work, adapt settings, or switch models before hitting a hard stop.