Welcome to Eema's Software
Tech to make your life easier
About
Never get surprised by an AI bill again.TokenWatch AI is a privacy-first usage dashboard that sits quietly in your
browser and tracks exactly how many tokens — and how much money — you're
spending every time you chat with ChatGPT, Claude, Gemini, or Perplexity.No API keys. No logins. No data leaving your device. Just a clean sidebar
that shows your real cost as you type.WHY TOKENWATCH AIMost people have no idea what a single ChatGPT or Claude conversation
actually costs until the subscription bill (or API invoice) arrives. AI
costs scale with usage, but the usage itself is invisible — until now.
TokenWatch AI estimates input and output tokens locally as you chat and
converts them into real dollar costs, using rate cards you can fully
customize per provider.KEY FEATURES📊 Live Usage Sidebar
A sliding dashboard overlays directly on chatgpt.com, claude.ai,
gemini.google.com, and Perplexity — showing input tokens, output tokens,
and estimated cost for your active conversation in real time.💰 Custom Cost Tracking
Set your own input/output rate per 1M tokens for each provider, switch
between USD, EUR, GBP, AUD, or JPY, and watch today's, this week's, this
month's, and lifetime spend update automatically.📈 Trend Dashboard & Month-End Projection
Visual daily/weekly/monthly cost charts plus a projected month-end total
based on your last 30 days of usage — so you can catch a runaway habit
before the invoice does.⚡ Smart AI Guard & Prompt Efficiency Score
Get a real-time efficiency score on your prompts, with detected "fluff"
flagged and one-click suggestions to trim wasted tokens before you hit
send.🧪 Sandbox Tokenizer Playground (Draft Lab)
Paste or draft any prompt and see its token count and cost instantly,
before you ever send it to a paid model.📜 Searchable History & CSV Export
Every conversation is logged with input/output tokens and cost, filterable
by provider and search term, and exportable to CSV for your own records,
budgeting, or expense reports.PRIVACY BY DESIGNTokenWatch AI runs 100% locally. There are no telemetry servers, no
remote databases, no analytics scripts, and no tracking APIs. Every token
count, every rate, and every history entry stays inside your browser's
own sandboxed storage — never transmitted anywhere.WHO IT'S FOR- Heavy ChatGPT Plus, Claude Pro, Gemini Advanced, or Perplexity Pro users
who want to see where their usage actually goes
- Freelancers and consultants who bill clients based on AI-assisted work
- Developers and prompt engineers optimizing token efficiency
- Anyone who's ever been surprised by how fast AI usage adds upGETTING STARTEDInstall TokenWatch AI, pin it to your toolbar, and open ChatGPT, Claude,
Gemini, or Perplexity. The sidebar appears automatically and starts
estimating tokens and cost as you chat — no setup, no sign-up, no API key.TokenWatch AI includes a free trial period, with lifetime premium access
available as a one-time purchase (no subscription) via secure, third-party
checkout through ExtensionPay/Stripe.Track what you spend. Trim what you waste. Stay in control of your AI costs.
PrivacyCore Principle: On-Device Only. TokenWatch AI reads the text of your prompts and responses on ChatGPT, Claude, Gemini, and Perplexity in order to estimate token counts and cost, and stores that text locally in your browser so you can review your usage history. That data is never transmitted to TokenWatch AI's developer, to any analytics or advertising service, or to any other third party. It stays on your device until you delete it.
1. What We Read and StoreTokenWatch AI's content script reads the text of each prompt you send and each response you receive on ChatGPT, Claude, Gemini, and Perplexity. It uses that text for two things, both performed entirely on your device:Estimating token counts and cost using a simple character-based heuristic (approximately one token per four characters).
Saving the prompt text, the response text, the computed token counts, and the computed cost to your browser's local storage (chrome.storage.local). This is what powers the searchable History log and the Analytics dashboard.
This stored text never leaves your device. It is not sent to TokenWatch AI's developer, and it is not sent to any analytics, advertising, or other third-party service.2. Where This Data LivesYour prompts, responses, computed token counts, and cost history are logged in your browser's local storage (via the chrome.storage API) and nowhere else. TokenWatch AI does not use background analytics trackers, web telemetry hooks, or remote database synchronization of any kind. When you clear your history from Settings, or uninstall the extension, 100% of this locally stored data, including the stored prompt and response text, is permanently deleted.3. Our Use of ExtensionPay (Payments)TokenWatch AI's premium upgrade ($9.99, one-time) is handled by ExtensionPay (extensionpay.com), a third-party payment and licensing service. The only network requests TokenWatch AI makes go to ExtensionPay, using an anonymous per-install identifier to check your trial and paid status. Your prompts, responses, and usage history are never included in these requests.If you purchase premium, or use "Restore Purchase" to activate premium on another device, ExtensionPay returns account information, which can include your email address and payment details such as the amount and date paid, and TokenWatch AI stores that information locally so it can remember your paid status. Your card and billing details are entered directly on ExtensionPay's and Stripe's own checkout pages, which TokenWatch AI never sees or has access to. See extensionpay.com for ExtensionPay's own privacy practices.4. Your Rights & OptionsYou maintain full control over this data: you can export your token and cost history to a CSV file at any time (this export includes usage numbers and costs, not the stored prompt or response text), clear your logs manually from Settings, or uninstall the extension, which deletes all locally stored data along with it.
Contact
Coming Soon