TokenTracker
Local-first AI token usage & cost tracker for 31 coding tools incl. Claude Code, Codex, Cursor, Gemini & DeepSeek Harness—with native apps. Never reads prompts.
What is TokenTracker?
Local-first AI token usage & cost tracker for 31 coding tools incl. Claude Code, Codex, Cursor, Gemini & DeepSeek Harness—with native apps. Never reads prompts.
Overview of TokenTracker
What it is
TokenTracker is a local-first, open-source token usage and cost tracking tool designed for developers using AI coding assistants. It monitors 46 different AI coding tools, including Claude Code, Codex, Cursor, Gemini, DeepSeek Harness, and many others. The software runs entirely on the user's machine, parsing log files to track token counts, timestamps, and model names. It never reads prompts, completions, or code. TokenTracker provides a local web dashboard, optional native desktop applications for macOS, Windows, and Linux, and includes features like a desktop pet, four native widgets, and 15 achievement tracks. The project is licensed under MIT and is available on GitHub.
Key features
- Monitors token usage for 46 AI coding tools out of the box, including Claude Code, Codex CLI, AStudio, Cursor, Gemini CLI, Antigravity, Kiro, OpenCode, OpenClaw, Every Code, Hermes Agent, GitHub Copilot, Kimi Code, CodeBuddy, WorkBuddy, WorkBuddy AI, Grok Build, oh-my-pi, OmO, pi, Dots, Prime Agent, Craft Agents, Reasonix, Muse Code, Kilo CLI, Kilo Code, Roo Code, Zed Agent, Goose, Droid, Mimo Code, DuMate, ZCode, Qoder, AnythingLLM Desktop, Claude Science, DeepSeek Harness, TRAE Work CN, LM Studio, Unsloth Studio, Devin CLI, Cline, MiniMax Code, Command Code, and TRAE.
- Local-first: runs on the user's machine, parsing logs locally with no account or API keys required.
- Zero config: hooks auto-install on first run; dashboard is available at localhost:7680 within 30 seconds.
- Native desktop apps: macOS menu bar app (with widgets), Windows system tray app, Linux AppImage (tray app), plus .deb and .rpm packages.
- Cost engine: prices over 2,200 models via LiteLLM (daily auto-refresh) with curated overrides for niche tools. Offline snapshot included.
Use cases
TokenTracker is useful for developers who want to monitor and analyze their usage of multiple AI coding assistants in one unified dashboard. It helps track token consumption and associated costs, visualize usage trends, and understand which models are being used most. The optional leaderboard allows comparing usage with other developers worldwide. The cross-device account view (optional cloud sync) merges usage from multiple machines. The Skills tab lets users browse and sync public skills to supported AI tools.
Getting started
Requirements: Node.js 20+ (CLI runs on macOS, Linux, Windows; native desktop app ships for macOS, Windows, and Linux). bash npx tokentracker-cli First run installs hooks, syncs data, and opens the dashboard at http://localhost:7680. For native desktop apps, download the appropriate installer from the GitHub releases page.
License and pricing
License: MIT. Pricing type: open source (free). The license and pricing are as stated in the repository metadata.
Based on the project's public README and repository metadata.