Nanako0129/TokenBar

AI token usage & quota monitor for the macOS menu bar — native Swift, Liquid Glass, 3D contribution graph. Tracks Claude Code, Codex, Cursor, OpenCode & 25+ agents locally.

What it solves

TokenBar provides a way for users to monitor their AI token usage and subscription quotas across multiple AI coding agents (such as Claude Code, Cursor, and Gemini CLI) from the macOS menu bar. It solves the problem of fragmented usage tracking by aggregating data from local session logs into a single, visible dashboard.

How it works

The application uses a hybrid architecture: a Rust-based engine (tokscale-core) handles the heavy lifting of parsing session logs, aggregating data, and calculating pricing, while a native Swift/SwiftUI frontend manages the macOS menu bar interface and dashboard views. It reads data on-device from local logs, meaning it requires no telemetry or external accounts to function.

Who it’s for

It is designed for developers and power users on macOS (specifically Apple Silicon Macs running macOS 14+) who use various AI coding agents and want to track their spending, token burn rate, and remaining subscription quotas in real-time.

Highlights

  • Multi-Agent Support: Tracks usage across 25+ AI coding agents.
  • Visual Monitoring: Displays today's tokens, costs, and quotas as signal bars, rings, or a "melting popsicle" in the menu bar.
  • Detailed Dashboard: Includes seven different data lenses (Models, Monthly, Daily, Hourly, Stats, and Agents) and an orbitable 3D usage graph.
  • Privacy-Focused: Operates locally on-device with no telemetry or account requirements.
  • Interactive Elements: Features a "sprinting cat" animation that speeds up as token usage increases.

Related

  • Project
    Lap-Platform/LAP

    A compiler and registry that transforms verbose API specifications into compressed, agent-native formats to reduce LLM hallucinations and token costs.

  • Project
    loLollipop/team-manage-refresh

    An integrated operational workbench for managing ChatGPT Team accounts, automating account import, seat distribution via redemption codes, and warranty management.

  • Project
    SakuraByteCore/codexmate

    A local-first control plane and dashboard for managing multiple AI coding agents, providing unified session browsing, provider bridging, and task orchestration.

  • Project
    agent-network-protocol/anp

    A multi-language SDK and reference implementation for the Agent Network Protocol (ANP), enabling AI agents to discover, authenticate, and communicate securely via standardized interfaces.