Token Tracker is a free, open-source, local-first dashboard for tracking token usage in AI coding agent CLIs. It gathers token counts in one place across 13 tools, giving a single view of model and project usage while recording only token counts rather than prompts or conversation content.
After installation with npx tokentracker-cli, the CLI auto-detects installed AI coding tools, adds the appropriate hooks or log readers, and opens a local dashboard at http://localhost:7680. Supported tools include Claude Code, Codex CLI, Cursor IDE, Gemini CLI, OpenCode, OpenClaw, Every Code, Kiro, Hermes Agent, GitHub Copilot, Kimi Code, oh-my-pi (omp), and CodeBuddy. The dashboard runs on the user's machine and does not require an account. Cloud sync is optional and is used only for the public leaderboard.
The local app presents a unified usage dashboard with input, output, cached, and cache-creation tokens across all 13 tools. It also includes per-model cost analysis across 70+ models, a usage-limits view with live rate-limit and subscription status for Claude, Codex, Cursor, Gemini, Kiro, and Antigravity, project attribution, activity heatmaps and trend charts, a skills browser, and a macOS menu bar app plus desktop widget for at-a-glance viewing. Token Tracker normalizes nine distinct log formats into a unified schema and serves aggregated bucketed counts from a local HTTP API on port 7680.
It is built for developers, founders, and engineering teams that rely on multiple AI coding assistants and need a single place to understand token consumption. The project is described as open source, with source on GitHub and an npm package named tokentracker-cli.
In the LLM eval & observability space, Token Tracker takes a focused approach. It focuses on tracking and analyzing token usage and costs across multiple AI coding agent CLIs locally and privately. Token Tracker is an open-source project aimed at developers using AI coding tools and agent CLIs. The project is open source (MIT). It runs on the web, the command line, macOS, Windows, and Linux.
mm7894215 builds and maintains Token Tracker, and it first shipped in 2026. Development happens publicly on GitHub with 794 stars and 508 commits in the last 90 days. Key capabilities include token usage tracking, cost analysis, and usage limits view.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do