Experience Win-CodexBar, a Windows system tray utility by Finesssee that monitors AI coding assistant usage and quotas. The app aggregates real-time usage, session and weekly consumption bars, reset timers, and provider-specific tabs for OpenAI Codex and Claude Code directly in the taskbar. Additional capabilities include local spend logging, dynamic tray icons, and secure cookie extraction. It targets developers and power users who need continuous, local telemetry and privacy-minded quota visibility for their AI tools.

Win-CodexBar brings native quota visibility to the Windows taskbar

The app is a native Windows port of the macOS CodexBar and displays live consumption data without opening a browser. It presents session and weekly usage bars, reset countdown timers, and separate provider tabs for OpenAI Codex and Claude Code. The frontend is built with React on Tauri and the backend is written in Rust, an architecture that separates UI rendering from the data-collection process.

It processes usage locally and minimizes external data exposure

The app uses local log parsing and secure cookie extraction to gather quotas and statistics, which supports the developer’s privacy-first design. The tool does not store passwords persistently and processes data on-device, and it is open-source under an MIT license so users can inspect the code. System requirements include WebView2 and modern 64-bit Windows 10 or Windows 11 environments.

Authentication and provider coverage suit developer workflows, with one notable limit

The app authenticates by extracting browser cookies, CLI logs, or OAuth tokens so it can show live stats without manual dashboard checks. That approach reduces dashboard switching for power users, and dynamic tray icons provide quick visual cues when quotas climb. In its current state, primary support focuses on OpenAI Codex and Claude Code, with parity for additional providers listed as an ongoing goal.

Practical choice for developers who need constant quota awareness

Win-CodexBar is a practical option for developers and researchers who require continuous, local visibility into AI usage and spend, backed by an open-source codebase and local processing. The main trade-off is its present provider scope, which centers on Codex and Claude rather than a broad set of services. For anyone needing unobtrusive taskbar telemetry and inspectable code, the app performs reliably within that scope.