GUIDE
How to check Claude Code usage on Mac
You can check how much Claude Code usage you have left two ways: ask Claude Code itself inside the terminal, or keep both limits visible in the menu bar so you never have to ask. Here's how each works.
Inside the terminal
Claude Code knows your own usage. Run /status in a session and it reports where you stand, including how close you are to a limit. It works, but it’s a pull, not a push: you have to remember to ask, in the right window, and the answer is a snapshot that’s stale the moment you keep working. For the two limits behind that number, see Claude Code rate limits, explained.
What you’re actually looking for
The two numbers that matter before a long run are the same two limits Claude Code enforces: how much of your 5-hour rolling window is left, and how much of the longer 7-day cap remains. Either one can be the thing that stops your agent, so a genuine “can I start this run?” check means reading both at once.
In the menu bar, always visible
NapCat surfaces both without a command. It borrows Claude Code’s own login - a Keychain prompt the first time - and fills your 5-hour and 7-day windows as two live bars in the Mac’s notch. A glance up tells you whether there’s room for a big autonomous run before you start it, rather than finding out when Claude Code stops halfway through.
Honest about how it works
- The usage meter reads your existing Claude Code login locally; the first read triggers a macOS Keychain prompt.
- It is optional - one toggle turns it off - and it is the only thing NapCat ever sends over the network. No telemetry otherwise.
- It sits alongside the live session cards, so usage and what your agents are doing share one glance.
Questions
How do I check Claude Code usage from the terminal?
Run /status inside a Claude Code session. It reports your current usage and how close you are to a limit. It's a manual check, so you have to remember to run it in the right window.
Which usage limits should I watch?
Two: the rolling 5-hour window and the longer 7-day weekly cap. Either one can stop an agent mid-run, so check both before starting a big autonomous session.
How does NapCat show my Claude Code usage?
It borrows Claude Code's own login (a Keychain prompt the first time) and fills your 5-hour and 7-day windows as two live bars in the notch, so both limits stay visible without a command.
Is the usage meter private?
It reads your Claude Code login locally and is the only network call NapCat makes. The meter is optional and one toggle turns it off. NapCat has no telemetry.
Related
Read Claude Code rate limits, explained, or see the full Claude Code support page.
NapCat does all of this from your Mac’s notch - for five dollars, once.
Download - free 2-day trial