Usage and limits

See the wall before you hit it.

Watch a sample session use up its 5-hour window, with a warning before it runs out.

Usage and limits · sample sessionSample data

5-hour window used

100 %

Session time

3 h 40 m

Heads-up at 80 %: about 45 minutes left at this pace.

  • SessionYour 5-hour window, and how fast you are using it.
  • WeeklyThe all-models weekly limit, with when it resets.
  • Heads-upA warning before the wall, not after.
Every service

Not only Claude. Credits, plans and GPU pods too.

Usage and limits · every serviceSample data
This 5-hour window62 %Limit reached about here, in 2 h 10 m
Document API credits69 %6,940 of 10,000 credits · no burn in the last hour
Tracing plan, free22 %220 of 1,000 events · $0 a month · resets on the 1st
GPU pods$0/hNone running · an idle pod is flagged within 10 minutes
How it works

How the runway works. From your usage to a heads-up.

  1. 01Earlyman reads your usageYour 5-hour window and your weekly limits are read and refreshed, each with where it came from and how fresh it is.
  2. 02It works out your runwayAt your current pace, it shows roughly how long you have left and when each window resets.
  3. 03You hear about it earlyA heads-up arrives before the wall, with a suggestion such as moving to a smaller model.
Try it

Watch the gauges and the week.

The real layout, with sample data. Nothing you click is sent or saved.

Earlyman · HomeInteractive demo · Sample data

Good morning. 4 things need you. Most urgent: a stuck agent is still costing money.

5-hour session62%heads-up at 80 % · resets in 1 h 40 m
This week · all models38%resets Sunday
This month · all vendors$412 of $600Claude Code, API keys, GPU pods
Needs you4 open
  • An agent has made no progress for 2 hA tool call is still pending. Stop it or start it again.$3.10 / h
  • PR #214 broke a rule: force push to mainBlocked before it ran. The agent carried on.1 rule
  • Search subagents run on the largest model418 of 483 this week. A smaller model is enough for search.−$38 / week
  • This session has run for 9 hHand off to a fresh session to stop paying for old context.−$6 / day

Questions

Which limits does it show?

The 5-hour session window and the weekly limits, each with its reset time.

Does it slow my agent down?

No. Earlyman only watches. It never sits between your agent and the model unless you choose the optional relay.

Can I get the warning somewhere else?

Yes. Alerts can go to a Mac notification, Slack, a webhook or your own script, once you add one.

Know your limit before it knows you.