If you build anything with the OpenAI or Claude APIs, you probably know the pattern.
You’re experimenting with prompts, testing scripts, or debugging an agent. Everything works fine, but there’s one thing that’s always a little unclear while you're developing:
How many tokens did that just use?
Most of the time you only see token usage later in logs or dashboards. That’s useful for analytics, but it’s not very helpful while you're actively coding.
During development I kept running into the same situations:
• testing prompt changes
• running quick API experiments
• building LLM tools
• debugging responses
And every time I wanted a simple answer in the moment:
“How many tokens did that call just use?”
So I built something extremely simple.
It’s a small macOS menu bar utility that shows token usage in real time while you're working. No dashboards or heavy analytics tools. Just a tiny counter in the menu bar so you can see usage immediately.
The goal was basically:
see token usage the same way you see CPU or battery status.
Nothing complicated, just immediate visibility while building.
If you're curious or want to try it, you can see it here:
I’d also be interested to hear how other people monitor token usage during development. Do you rely on logs, dashboards, or something else entirely?