Auto-collect token usage from 20+ AI coding tools, aggregate it locally, and read real cost trends in one dashboard. No account or API key required to start — just one command.
⭐ If TokenTracker saves you time, please star it on GitHub.
You can — that is the honest alternative, and for a single tool it is enough. TokenTracker earns its place once you use more than one:
- One number instead of six tabs. Claude, Codex, Cursor, Gemini and Copilot each bill in their own dashboard, on their own reset schedule, in their own units. Nobody adds them up for you.
- Subscriptions hide the number entirely. A flat monthly plan shows you a quota bar, not what your usage would have cost. TokenTracker prices every token against public model rates, and you can enter what you pay to see the two side by side. The figure is list-price-equivalent for the window on screen, not a bill you would have paid — and when some usage cannot be priced it says so rather than quietly reporting a smaller number.
- Per-project and per-model, not just per-account. Billing pages answer "what do I owe this month". This answers "which repo, which model, and which hour" — the resolution you need to actually change something.
- Quota chips before you hit the wall. Live plan-limit usage sits on each provider's card, so a 5-hour window running out is something you see rather than something you discover.
If you only use one tool and never care about per-project cost, the provider's own page is genuinely fine. This is for the rest.
Requires Node.js 20+.
npx --yes @ipv9/tokentracker-cliThat's it. The first run installs hooks, syncs your data, and prints the dashboard URL, usually http://127.0.0.1:7680. Pass --open if you want TokenTracker to ask the OS to open your browser.
Prefer a global install for shorter commands:
npm install -g @ipv9/tokentracker-cli
tokentracker # start the dashboard and print the URL
tokentracker serve --open
tokentracker sync # sync now
tokentracker status # check which tools are connected
tokentracker doctor # health checkIf you'd rather not keep a terminal open, both native apps are on the Releases page:
| Platform | Download | What it adds |
|---|---|---|
| macOS 12+ | TokenTrackerBar.dmg |
A menu-bar app — live token count in the menu bar, launch at login, sync and update from a click, plus a desktop widget. |
| Windows | TokenTracker-Setup.exe |
The same dashboard as a standalone app. |
Both bundle their own Node runtime, so there is nothing else to install. They share the same local data as the CLI — run either, or both. The desktop builds are cut less often than the npm package, so the latest release tag usually trails the npm version badge above.
- 🔒 Your usage data never leaves your machine. Token counts and timestamps only — never prompts, responses, or file contents. No account, no telemetry, no analytics, no phone-home. TokenTracker does make a few outbound calls on your behalf (model prices, your own plan quotas); every one is named in Privacy below, and none of them carry your usage.
- 📊 One calm web dashboard. Your whole picture in the browser at a local URL, no login — light or dark, auto-refreshing while the tab is open. What's on it ↓
- 📈 Quota at a glance, on every card. Live plan-quota usage (e.g. 5h + weekly) as color-coded chips right on each provider's card — see how close you are to your limits without leaving the overview. Where the provider reports countable units you get the actual number rather than a percentage to convert in your head: GitHub Copilot reads
158/300premium requests. Full windows + reset countdowns on the Limits page. Covers Claude, Codex, Cursor, Gemini, Kimi, Z.AI, Kiro, Copilot, and Antigravity. - 💰 Cost you can trust — and a price tag when it can't. 2,200+ models priced from LiteLLM (refreshed daily) with a bundled offline snapshot, so USD totals are right even without a network. A model too new to have a price is badged pricing missing rather than quietly counted as $0, and prices refresh in the background instead of waiting for a restart. Cross-provider records are de-duplicated to match each provider's own billing.
- 🔌 20+ tools, zero config. Claude Code, Codex, Cursor, Gemini, Copilot, Antigravity, OpenCode, Kiro, Zed, Goose, and more — auto-detected, hooks auto-install on first run. Zero to dashboard in ~30 seconds.
- 🧩 Skills tab. Syncs 250+ public skills across your tools.
| Dark | Light |
|---|---|
|
 |
On one screen:
- Total tokens + cost for the selected window (24h / day / 7d / 30d / total / custom), with a browser-timezone "Updated" stamp.
- Usage Trend — token volume over time, right beside the total.
- Provider breakdown — every tool's share and top models, with each provider's live quota chips (e.g. 5h + weekly — a count like
158/300where the provider reports one, otherwise a percentage) on its card and a click-to-expand per-model drill-down. - Context breakdown — where Claude's and Codex's tokens actually go (input / output / cache / reasoning), shown side by side.
- Activity heatmap — daily usage at a glance.
- Project & daily tables — per-project totals and a day-by-day breakdown with cost and $/MTok.
Auto-refresh runs only while the tab is visible (Off / 30s / 60s / 120s, default 30s).
Auto-detected on first run — no manual plugin or hook wiring:
Claude Code · Codex CLI · Cursor · Gemini CLI · GitHub Copilot · Antigravity · Kiro · OpenCode · OpenClaw · Every Code · Hermes · Kimi Code · CodeBuddy · Grok Build · Droid · oh-my-pi · pi · Craft Agents · Kilo CLI · Kilo Code · Roo Code · Zed Agent · Goose
Each tool is connected one of three ways, all automatic: a SessionEnd/notify hook (Claude Code, Codex, Gemini, Every Code, CodeBuddy, Grok Build), a bundled plugin linked via the tool's own CLI (OpenCode, OpenClaw), or a passive reader that only reads files the tool already writes — SQLite, JSONL, OTEL exports (Cursor, Kiro, Copilot, Zed, Goose, and the rest).
Run tokentracker status to see each integration's state. Missing your tool? Open an issue — a new provider is usually one parser file away.
Rate-limit providers are auto-detected where possible. For Z.AI / GLM Coding Plan quota windows, export ZAI_API_KEY or ZHIPU_API_KEY before starting the local server. If Claude-compatible tooling is pointed at Z.AI with ANTHROPIC_BASE_URL=https://api.z.ai/... and ANTHROPIC_AUTH_TOKEN, TokenTracker can reuse that auth token for the local limits endpoint.
AI CLI tools → hooks / passive readers → local queue file → dashboard
(logs) (token counts only) (30-min buckets) (your browser)
- Your AI tools write logs during normal use.
- Lightweight hooks (or passive file readers) pick up token counts locally — never prompt or response content. Some tools keep their logs in SQLite (Cursor, Kiro, Zed and friends); TokenTracker only ever reads those.
- Counts are aggregated into 30-minute UTC buckets and appended to one plain-text file:
~/.tokentracker/tracker/queue.jsonl. - The dashboard reads that file and renders it in your browser's timezone.
No account, no upload of your usage, and no server to sign in to.
| Protection | What it means |
|---|---|
| No content | Only token counts and timestamps. Never prompts, responses, or files. |
| Your usage stays local | Every count TokenTracker collects is written to one file on your disk and read back by a server on your own machine. There is no endpoint it uploads usage to. |
| Auditable in one command | You don't have to take our word for it — the store is an append-only text file you can open yourself: cat ~/.tokentracker/tracker/queue.jsonl. It's numbers and timestamps. |
| No telemetry | No analytics, no crash reporting, no phone-home, no account. |
Outbound calls. TokenTracker is local-first, not network-free. It reaches these hosts and no others. None of them carry your usage data — but some do reveal that you are asking, so they are listed with who makes the call.
This table is checked in CI against outbound-hosts.json: a host the code can reach but the file does not declare fails the build, and so does a declared host missing from this table. It is not maintained by memory.
| When | Host | From | Why |
|---|---|---|---|
| Pricing refresh (daily) | raw.githubusercontent.com |
server | The public LiteLLM price list. Anonymous — no credentials, nothing sent. Works offline from a bundled snapshot. |
| Quota chips + Limits page | api.anthropic.com, chatgpt.com, api.openai.com, cursor.com, www.cursor.com, cloudcode-pa.googleapis.com, api.kimi.com, api.z.ai, api.github.com |
server | Asks your provider about your plan limits, using credentials already on your machine. Only for providers you actually use. |
| Token refresh | auth.openai.com, oauth2.googleapis.com, auth.kimi.com |
server | Renews those same provider credentials when they expire. |
| Skills tab — search | skills.sh |
server | Searches the public skills directory. Sends the terms you typed. Only when you open that tab. |
| Skills tab — install | api.github.com, raw.githubusercontent.com |
server | Reads a public repository's file tree and downloads the skill's files. Carries the repository owner, name, branch and path — no usage data. |
| Currency conversion | open.er-api.com |
browser | Public USD exchange rates. Anonymous. Only when you pick a non-USD currency. |
| Star count in the header | api.github.com |
browser | The star count for this project's own repository — a fixed public URL that says nothing about you. |
| IP check page | ip.net.coffee, d.ip.net.coffee, 1.1.1.1, claude.ai, www.anthropic.com |
server + browser | Only if you open that page, whose entire purpose is showing you your own IP, your DNS resolver, and whether Anthropic is reachable. The d. subdomain is a per-run DNS probe: your browser resolves a unique name so the page can see which resolver answered. |
| IP check page — VPN leak test | stun.l.google.com (Google), stun.cloudflare.com |
browser | WebRTC STUN binding requests, which is how the page detects whether a VPN leaks your real address — and which necessarily discloses that address to the STUN server. This is the one place the dashboard talks to Google. Only if you open that page. |
npx startup |
registry.npmjs.org |
server | How npx works — it downloads the package. A global install avoids it. |
What is deliberately absent: the Projects panel does not fetch repository avatars or star counts. Doing so would have put the name of a repository you have checked out — a private one included — into a URL sent to GitHub from your browser. It did, until #100. Project rows now render a local icon.
Most users never touch this — defaults are sensible.
| Variable | Description | Default |
|---|---|---|
TOKENTRACKER_DEBUG |
Debug output (1 to enable) |
— |
TOKENTRACKER_HTTP_TIMEOUT_MS |
HTTP timeout (ms) | 20000 |
CODEX_HOME |
Override Codex CLI directory | ~/.codex |
GEMINI_HOME |
Override Gemini CLI directory | ~/.gemini |
TOKENTRACKER_GROK_HOME |
Override Grok Build directory | ~/.grok |
To force a dashboard port, use your shell's environment-variable syntax (otherwise TokenTracker auto-picks the next free port from 7680):
PORT=7700 npx --yes @ipv9/tokentracker-cli serve$env:PORT = 7700
npx --yes @ipv9/tokentracker-cli serveBrowser auto-open is opt-in: tokentracker serve --open. Background services and headless shells should use the default no-open behavior and open the printed URL manually.
If you want the dashboard up all the time but don't want the desktop app, the repo ships a launchd installer. It registers two LaunchAgents — the dashboard on port 7680, and a periodic background sync — both pinned to a specific published version:
git clone https://github.com/pitimon/TokenTracker.git
cd TokenTracker
./scripts/install-local-service.sh # remove later with ./scripts/uninstall-local-service.shmacOS only; it uses launchd directly. On Linux, the same effect is a small systemd user unit running tokentracker serve --sync --no-open.
git clone https://github.com/pitimon/TokenTracker.git
cd TokenTracker && npm install
npm run dashboard:build && node bin/tracker.js
npm run ci:local # the full gate: build + tests + validatorsSetup details, the test layout, and how to add a new tool integration are in CONTRIBUTING.md. Source-backed engineering documentation starts at openwiki/README.md.
A tool isn't being detected
tokentracker status # see each integration's state
tokentracker doctor # deeper health checkIf a tool you use shows as not configured, run tokentracker init — it re-runs detection and installs anything missing. Still missing? Open an issue with the doctor output.
Port 7680 is already in use
The server auto-picks the next free port (7681, 7682, …) and logs it on startup. To force one, use the Bash or PowerShell command in Configuration. To see what's holding port 7680:
lsof -i :7680Get-NetTCPConnection -LocalPort 7680 -ErrorAction SilentlyContinueLinux: spawn xdg-open ENOENT
TokenTracker does not require xdg-open to run. The dashboard URL is printed on startup, so you can open it manually from another browser.
If you explicitly pass --open on a desktop Linux machine and see spawn xdg-open ENOENT, install the desktop opener package:
sudo apt update
sudo apt install -y xdg-utilsOn servers, SSH sessions, CI, systemd units, and other headless environments, leave browser opening disabled and run:
tokentracker serve --sync --no-open
tokentracker doctorWindows / PowerShell: start TokenTracker with npx
For an interactive PowerShell session:
npx --yes @ipv9/tokentracker-cli serve --openFor headless PowerShell, CI, or a background process, keep browser opening disabled:
npx --yes @ipv9/tokentracker-cli serve --no-sync --no-openTokenTracker never stops another process to free a port. When no port is specified it tries the next one up (7681, 7682, …); an explicit --port or $env:PORT that is already in use fails startup and prints an alternative command to run.
Remove everything
tokentracker uninstallRemoves every hook TokenTracker installed across all detected tools, plus local config and data. Safe to re-run.
One thing it does not touch: if you set up the always-on macOS service yourself with scripts/install-local-service.sh, that LaunchAgent is installed outside the CLI and keeps restarting the dashboard. Remove it first:
./scripts/uninstall-local-service.sh(The CLI never installs a LaunchAgent, so if you have only ever run npx/tokentracker, there is nothing extra to clean up.)
- Bugs / features — open an issue
- Security — see SECURITY.md (please don't file public issues for security reports)
- Pull requests — see CONTRIBUTING.md
The Clawd character design belongs to Anthropic. This is a community project with no official affiliation with Anthropic.