AI USAGE MONITOR / MAC
Know what’s left in Codex and Claude Code, when it resets, and how to make your next task count.
Free · No credit card · macOS 13+ · Compatible AI account required
THE APP, AT A GLANCE
Weekly limits, five-hour usage and reset times. Together.
Explore the example. Your account readings stay on your Mac.
Your ambition has no limits. Your plan does.
Data through 12 Sep, 01:00 UTC
Selected day
2026-09-12 · 4,000 tokens · partial day
Color intensity shows activity. Neutral squares have no reading; select a day for details.
Tokens measure how much text the AI processes. These are not credits or money. Days without data stay empty.
ChatGPT · Chat tab · GPT-6
Create an interactive task planner from the tasks I paste below. Let me change priority and estimated time, then show a realistic order for a two-hour work session. Keep my task names, flag missing estimates and explain the ordering in three lines. If interactive UI is unavailable, give me a simple table instead. Tasks: [paste your tasks].
Codex for Mac · in the composer
Start with these 5.
Outline the task before acting.
Inspect code changes for problems.
Inspect context and rate limits.
Open usage settings when available.
Condense this chat’s context.
Checking official sources…
No daily data received
Selected day
2026-09-12 · No data
Color intensity shows activity. Neutral squares have no reading; select a day for details.
Claude reports session and weekly limits, but not daily token totals to this app. No annual usage is inferred from those percentages.
Claude Code or Codex · one reproducible bug
Help me diagnose one reproducible bug before changing code. Observed: [what happens]. Expected: [what should happen]. Reproduction: [smallest steps or mocked input]. Relevant project or files: [paths, if known]. Keep unchanged: [constraints]. Read the relevant project instructions and inspect the smallest useful set of files. Give me a five-line brief: objective, observed evidence, actual relevant paths, unknowns, and the smallest check that can reproduce this failure. Mark paths and causes UNVERIFIED until inspected; do not invent a test command. For the check, list setup, action and observable pass condition. Include the reported failure and one retry or boundary case. Separate a mock result from a real account result. If you cannot run it, mark it NOT RUN and name what is missing. Do not edit during this diagnostic step. Stop with the proposed scope and check. Once I authorize a fix, change only what the evidence supports, rerun the relevant check, and report the actual command/output or manual observation. Passing one test does not prove the whole product works. Do not publish or change credentials or permissions. Before a long follow-up, check remaining plan capacity and reset time in INMC or the provider. An unavailable reading is unknown; tokens, plan capacity and API cost are different.
Claude Code · in the terminal session
Start with these 5.
Outline a task before changes.
Inspect usage and plan limits.
See what occupies the context window.
Summarize this session to continue.
Reset context for unrelated work.
Checking official sources…
Your AI usage stays on this Mac. The percentage is plan capacity, not money.
Illustrative data, not readings from your accounts.
YOUR AI TOOLKIT
Codex and compatible Claude Code accounts. Current readings on your Mac, with no browser extension.
ChatGPT plan usage on your Mac
Check Codex usage and limitsCompatible Claude Code accounts
Check Claude Code usage and limitsAMBITION, WITH INTENTION
Check your limits. Brief the task. Keep creating.
Check available capacity and reset times before a bigger task.
Copy into your AI chat. Adapt the details to your task.
Help me plan a focused 45-minute work session.
Tasks: update my homepage, write one follow-up email, and outline three social posts.
Choose one deliverable to finish first. Explain the choice and give me three concrete steps. Keep the other tasks for later. Do not estimate AI quota or promise these tasks will fit my remaining usage.
One more idea.
One better brief.
Name the outcome, context, constraints and what done means.
Copy into your AI chat. Adapt the details to your task.
Improve the webpage I share with you.
make the product and next step easier to understand.
keep the approved brand, original logo and real product colors. Do not invent features or claims.
the headline explains the product, there is one clear primary action, and the layout works on mobile.
Ask for the page if I have not provided it. Suggest the three highest-impact changes before editing.
Keep what works. Ask for a specific correction before starting over.
Copy into your AI chat. Adapt the details to your task.
Review the previous draft against my original brief.
List up to three specific mismatches. Explain the smallest correction for each, then revise only those parts. Preserve everything that already meets the brief.
Separate verified facts from assumptions. Ask if an essential detail is missing. Do not add unsupported claims.
Select a recorded day, compare it with your own recent average, and get a practical tip for your next task. Neutral days have no reading.
Claude does not report daily token totals to this app.
White marker = your average · Example data
For a routine edit, try a smaller scope and lighter reasoning, then verify the result.
Illustrative activity · stronger color means more activity
Tokens aren’t credits or money.
Open release notes and videos from OpenAI and Anthropic.
Platform news is separate from the version installed on your Mac.
Get useful notices for low capacity, recent pace and saved resets. You choose which alerts to keep.
Understand usage alerts →Open the DMG and double-click Install I Need My Credits.
Start with the tool you use. Codex uses your local sign-in; Claude Code connects only if you choose it.
Your usage and reset times appear in the menu bar.
Yes. The monitor needs no subscription or credit card. Your AI plans are separate.
Codex and Claude through Claude Code. Personal Claude accounts need Pro or Max for this connection; Team or Enterprise eligibility depends on the organization. No browser extension is required.
macOS 13 or later, on Apple silicon or Intel. Intel hardware testing is still pending. Windows is not supported.
No. It shows the usage included in your existing plans. Each AI provider sets its limits and resets.
No. AI usage readings stay on your Mac. Optional product reports require your consent. See Privacy for details.
