Analysis updated 2026-05-18
See total spending and a daily breakdown for using the Pi AI agent.
Check which programming languages and AI models you have used most in Pi sessions.
Pull Pi agent log files from a remote server over SSH instead of only reading local files.
| phun333/pi-infobar | wondaggvcb-oss/lazy-bear-desktop | kellyvv/openreshot | |
|---|---|---|---|
| Stars | 48 | 46 | 51 |
| Language | Swift | Swift | Swift |
| Setup difficulty | easy | moderate | moderate |
| Complexity | 2/5 | 2/5 | 3/5 |
| Audience | developer | general | general |
Figures from each repo's GitHub metadata at analysis time.
Because the developer lacks a paid Apple developer account, macOS flags it as unverified, the install command clears the security flag.
Pi Stats is a small macOS application that sits in your menu bar and shows you a dashboard of how much you have spent using the Pi AI agent. It displays your total spending, a breakdown by day, which programming languages you have written code in, which AI models you used, which projects you worked on, and how many tokens you consumed. All of this is calculated by reading session log files that the Pi agent saves to your computer. The app is local by default, meaning it never sends your data anywhere. It reads files from a folder on your Mac, processes them, and shows you the results. There is an optional remote mode where you can configure it to pull log files from your own server over SSH, but even then nothing is uploaded: data only flows from your server to your Mac. Installing it takes one extra step because the developer does not have a paid Apple developer account, so macOS will initially block it as unverified. The README provides a single Terminal command that downloads the app, installs it, clears the security flag that macOS sets on downloaded applications, and opens it. There is also a manual installation path if you prefer to download a disk image file instead. Once running, a small pi symbol appears in your menu bar. Clicking it opens a panel with tabs for cost, languages, models, projects, and token usage. You can filter each view by the last day, seven days, thirty days, or all time. The app is built with Swift and supports both Apple Silicon and Intel Macs running macOS 14 or newer. The source code is available under the MIT license, and build scripts are included for anyone who wants to compile it themselves.
Pi Stats is a macOS menu bar app that reads local log files to show how much you have spent, and on what, while using the Pi AI coding agent.
Mainly Swift. The stack also includes Swift, macOS.
Released under the MIT license, so you can use, modify, and redistribute it freely as long as you keep the copyright notice.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Verify against the repo before relying on details.