A little while ago, I posted here asking what you'd look for first in a redesigned CodexBar Lite.
The feedback was incredibly useful - especially around glanceability, keeping the app lightweight, and making sure a menu bar utility stays out of your way.
After using the redesign myself for a while and polishing the little things, it's now live.
The goal hasn't changed: keep CodexBar Lite minimal, privacy-first, and focused on doing one thing well.
I've just finished a pretty substantial UI refresh for CodexBar Lite and I'm using it myself before shipping it.
The functionality is largely the same. The focus this time wasn't adding features, it was making the app feel more polished while staying true to its original philosophy:
Minimal
Privacy-first
Native macOS
Before I hit release, I'm curious: If you were testing a menu bar app, what's the first thing you'd notice or judge?
CodexBar Lite closed at #13 Product of the Day with 97 points. For a first Product Hunt launch, I honestly couldn't have asked for more.
I want to thank everyone who supported the launch, whether you upvoted, commented, tried the app, shared feedback, or simply stopped by to take a look. Every interaction meant a lot.
The launch may be over, but the project certainly isn't.
CodexBar Lite is a native macOS menu bar app that keeps your OpenAI Codex usage visible at a glance.
It uses your existing Codex CLI session - no browser cookies, Keychain permissions, API keys, or third-party accounts required.
Monitor usage, reset times, and receive notifications, all from a lightweight, open-source app.
The moment a usage tracker asks for half a dozen permissions, something has already gone wrong.
I wasn't looking for another dashboard. I wasn't looking for analytics. I definitely wasn't looking to hand over Chrome access, keychain access or browser cookies. I wanted to answer one question.
"Can I keep coding, or am I about to hit my limit?"
One thing I've tried to avoid with CodexBar Lite is feature creep. Every feature starts as a good idea. Add enough of them and you've accidentally built another dashboard.
So before I add anything, I usually ask myself one question:
"Does this help someone get back to coding faster?"
