stormzhang

Track token usage across local AI agents (Claude Code, Codex) — CLI dashboard with cost analysis, rate limit monitoring, and session tracking

14
1
100% credibility
Found May 09, 2026 at 14 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Token Tracker is a dashboard tool that reads local data from AI coding agents like Claude Code and Codex to display token usage, equivalent costs, session details, and rate limit status in an interactive terminal view.

How It Works

1
🕵️‍♀️ Discover Token Tracker

While chatting with your AI coding buddies like Claude or Codex, you hear about a handy tool to watch how much you're using them.

2
📥 Set it up quickly

You grab the tool onto your computer with one easy download that sets everything up automatically.

3
🔍 Launch for the first time

Open your terminal and type a simple name, and it finds your AI tools and prepares to show their info.

4
📊 See the magic dashboard

A beautiful screen pops up with charts of your token usage, costs, sessions, and limit warnings all in one place.

5
Pick your view
📅
Daily/Weekly View

Check how much you used today or this week with costs and trends.

💬
Recent Sessions

See details on your latest chats like project, length, and tokens spent.

⏱️
Limits & Blocks

Watch rate limits with reset timers and active usage speeds.

Track like a pro

Now you effortlessly monitor spending, stay under limits, and make smarter use of your AI helpers every day.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 14 to 14 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is token-tracker?

Token-tracker is a Python CLI dashboard that monitors token usage for local AI agents like Claude Code and Codex, pulling data directly from their JSONL logs and SQLite stores. It shows real-time burn rates, equivalent costs via LiteLLM pricing, rate limit progress bars with reset timers, and breakdowns by session, day, week, or month—run `tt` for an interactive view or `tt claude` for agent-specific stats. For devs chaining AI tools, it solves the blind spot of creeping token costs and hitting 5h/7d limits without digging through files.

Why is it gaining traction?

Zero-config setup auto-detects agents and stays read-only, unlike manual log parsers or broader github track-activity tools. The Rich-powered terminal UI packs heatmaps, trends, and P90 projections into one screen, with hooks for live Claude status lines—beats exporting to spreadsheets. As a claude token tracker github alternative to crypto-focused ones like solana-token-tracker github or token tracker etherscan, its focus on AI session insights hooks power users tracking copilot-like local flows.

Who should use this?

AI engineers grinding Claude Code for codegen or Codex for local inference, especially those billing sessions or dodging rate walls. Local LLM tinkerers monitoring track github copilot usage equivalents, or teams splitting token tracker costs across projects. Skip if you're on cloud APIs without local logs.

Verdict

Grab it via `pip install token-tracker` if you're deep in these agents—solid features for 14 stars, but 1.0% credibility signals early days with thin docs and no tests. Watch for polish before production reliance.

(187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.