Launch
Claude Monitor
Visit
Example Image

Claude Monitor

Never get cut off by a Claude usage limit again

Visit

Claude Monitor shows how much Claude you have left before you hit a usage limit. A badge in your browser toolbar tracks your 5-hour session and weekly limit live, with a countdown to every reset and desktop alerts at 80% and 95%. One click opens the full picture: per-model weekly caps, extra-usage credits in real money, Claude Code routine runs and 30 days of history you can export. Turn on the optional bar and the limits sit right above claude.ai's chat box. It even works on the Free plan, where Claude stopped showing usage in August 2026. Free, open source (MIT), no account, and nothing leaves your browser.

Example Image
Example Image
Example Image
Example Image
Example Image

Features

  • Live 5-hour session and weekly limit in the toolbar badge
  • Countdown to every reset
  • Desktop alerts at 80% and 95% (configurable)
  • Per-model weekly caps and extra-usage credits in real money
  • Claude Code routine runs
  • Optional usage bar above the claude.ai chat box
  • Works on the Free plan
  • 30 days of history with CSV/JSON export
  • Open source (MIT), no account, no telemetry

Use Cases

  • Pace a long coding session in Claude Code without hitting the 5-hour limit
  • Know when your weekly limit resets before starting a big task
  • See your limits on the Free plan, where claude.ai no longer shows them
  • Track extra-usage spend so the credits do not surprise you

Comments

Hi Fazier, I'm Martín, the maker. I built Claude Monitor because I kept hitting Claude's 5-hour and weekly limits without warning, and the only way to check was reloading the settings page. It puts those numbers in the browser toolbar with a countdown to each reset and alerts before you run out. The latest version adds an optional bar right above the claude.ai chat box, and Free plan support: since August 2026 claude.ai shows no usage to Free accounts, so the extension reads the limits from each reply instead. It is free and open source (MIT), and nothing leaves your browser. I would love your feedback: what would make it more useful for how you use Claude?

The per-model weekly caps plus extra-usage credits shown in real money is a smart touch, since that's the part that usually surprises people mid-project. How do you read these numbers without an official usage API, and how do you plan to handle it if claude.ai changes its format? Great that it stays local with no telemetry.

Showing per-model weekly caps and the 80%/95% alerts right in the toolbar is a practical way to pace a long coding session. Since it's local-only, how does it read the usage data without an account or API key?

custom-img
The team behind FideNubi, a free and ind...

The countdown to the 5-hour reset is exactly what I miss during long Claude Code sessions. Does it track usage from Claude Code in the terminal as well, or only what happens in the claude.ai tab? The 80%/95% alerts are a nice touch.

getting cut off mid-task is my biggest pain with claude code. does it warn before you hit the limit or just track usage? a heads up like "~20 min left at this pace" would be gold

custom-img
OEM IoT hardware maker

The per-model weekly caps and the countdown to reset are the parts I would use most, since hitting a limit mid-task is frustrating. Keeping everything local with no account is a sensible choice for a tool that sits on top of a chat page. Would be nice to see a simple way to share the exported history in CSV form.

custom-img
I build & lead the engineering behind AI...

The Claude usage limit cutoff is real - teams get blocked mid-pipeline without visibility. Having alerts before you hit the ceiling saves both money and operational headaches. This solves the monitoring gap that Anthropic's dashboard doesn't cover.

custom-img
Building ClearedToShip

I've hit the 5-hour limit halfway through a Claude Code session more than once, so the 80% alert would've saved me. Does the badge count Claude Code usage from the terminal too, or only what happens in the browser?

You built what Anthropic didn't! Thank you

Premium Products

Comments

Hi Fazier, I'm Martín, the maker. I built Claude Monitor because I kept hitting Claude's 5-hour and weekly limits without warning, and the only way to check was reloading the settings page. It puts those numbers in the browser toolbar with a countdown to each reset and alerts before you run out. The latest version adds an optional bar right above the claude.ai chat box, and Free plan support: since August 2026 claude.ai shows no usage to Free accounts, so the extension reads the limits from each reply instead. It is free and open source (MIT), and nothing leaves your browser. I would love your feedback: what would make it more useful for how you use Claude?

The per-model weekly caps plus extra-usage credits shown in real money is a smart touch, since that's the part that usually surprises people mid-project. How do you read these numbers without an official usage API, and how do you plan to handle it if claude.ai changes its format? Great that it stays local with no telemetry.

Showing per-model weekly caps and the 80%/95% alerts right in the toolbar is a practical way to pace a long coding session. Since it's local-only, how does it read the usage data without an account or API key?

custom-img
The team behind FideNubi, a free and ind...

The countdown to the 5-hour reset is exactly what I miss during long Claude Code sessions. Does it track usage from Claude Code in the terminal as well, or only what happens in the claude.ai tab? The 80%/95% alerts are a nice touch.

getting cut off mid-task is my biggest pain with claude code. does it warn before you hit the limit or just track usage? a heads up like "~20 min left at this pace" would be gold

custom-img
OEM IoT hardware maker

The per-model weekly caps and the countdown to reset are the parts I would use most, since hitting a limit mid-task is frustrating. Keeping everything local with no account is a sensible choice for a tool that sits on top of a chat page. Would be nice to see a simple way to share the exported history in CSV form.

custom-img
I build & lead the engineering behind AI...

The Claude usage limit cutoff is real - teams get blocked mid-pipeline without visibility. Having alerts before you hit the ceiling saves both money and operational headaches. This solves the monitoring gap that Anthropic's dashboard doesn't cover.

custom-img
Building ClearedToShip

I've hit the 5-hour limit halfway through a Claude Code session more than once, so the 80% alert would've saved me. Does the badge count Claude Code usage from the terminal too, or only what happens in the browser?

You built what Anthropic didn't! Thank you

Premium Products