Privacy Policy
Last updated October 10, 2026. Effective October 10, 2026.
The short version. Model Stack (modelstack-ai.com) runs your tasks through AI models and bills you for them. We keep your account, billing records and anything you choose to save. Your task text is sent to the AI companies that do the work. Card numbers go to Stripe and never touch our servers. We don't sell your data.
Who we are
"Model Stack", "we" and "us" mean the business that runs modelstack-ai.com. You can reach us at modelstack44@gmail.com.
What we collect
| What | Details |
|---|---|
| Account | Your email address, business name, and a password. The password is stored only as a salted, deliberately slow hash, never in readable form. If you use "Continue with Google" or "Continue with Microsoft", we receive your name and email address from them, and nothing else. |
| Billing | Your Stripe customer reference, your card's brand and last four digits, what each task cost, your plan and its fees, any free credit you have left, and your payment history. Your full card number is entered on Stripe's own page and is never seen or stored by us. |
| Confirming your email | To give a new account its free credit, we email a six-digit code to the address on the account. We keep only a scrambled form of the code until it expires, and a note that the address was confirmed. |
| Messages to us | If you use "Talk to us" on the pricing page, we keep what you type there (your name, email address, company, team size and message) so we can reply, and we email it to ourselves. |
| Tasks | A short title, the type of work, the status, the time, which models were used and what it cost. Your chat history, including what you typed and the answers, is kept in your own browser rather than on our servers. |
| Things you save | Agents (their instructions), automations, scheduled tasks and their last result, projects, and the facts the memory feature remembers about your business. Only what you choose to create. |
| Files | Images you upload, and images, narration or video you generate with us. |
| Team | If you invite colleagues, their email addresses and the spending limit you set for them. |
| Connected accounts | If you link Google through Plugins, we store a token so a task can save a Google Doc for you later, along with the permissions you granted (see "Google user data" below). |
| Websites you let agents use | If you set up an agent to act on a website (for example your CRM or accounting site), you sign in yourself in a live browser window running on our servers. Your password and anything else you type there goes straight to that site's own page: it is not saved, and what you type is not logged. The browser keeps the session the site gives it (its cookies) on the service that runs our browsers, so agents can use the site as you; we record only that you've signed in. Disconnect on the Plugins page erases it. The Browser page and "Browse the web" use a separate, temporary browser on the same service: it has no sign-ins of its own, and everything in it (pages, cookies, anything you type) is deleted when it closes or goes idle. If you sign in to a site yourself in that window, that sign-in disappears with it. While an agent works, it sees the pages it visits as screenshots and text, which are sent to the AI companies behind the model doing the work, the same way a task's text is. |
| Technical | A sign-in cookie to keep you signed in, and your IP address, used for abuse protection and rate limits. |
How we use it
- To run your tasks and show you the results.
- To bill you, and to enforce spending limits.
- To keep your account secure, including password resets by email.
- To prevent abuse and to fix problems.
We don't sell your personal data, and we don't use it for advertising.
Who we share it with
We share only what's needed to provide the service:
- AI providers. To do your work, the text of a task (and any images you attach) is sent to the AI companies behind the models chosen for it, such as Anthropic, OpenAI and Google, sometimes through OpenRouter. What they do with it is governed by their own terms and privacy policies. A task is sent to several providers, because several models work on it.
- Stripe, which handles your card and payments.
- Our hosting and network providers, which run the site (currently Render, with Cloudflare in front).
- An email provider, used to send password reset codes and team invitations.
- Authorities, only where the law requires it.
Google user data
Model Stack uses Google in two separate ways, and you choose each one.
1. "Continue with Google" sign-in
- What we access: your name, your email address, and whether Google has verified that address (the basic
openid,emailandprofilepermissions, and nothing more). - Why: to create your Model Stack account, or sign you in to the one you already have.
- What we keep: that email address and name, as part of your account. No Google access token is kept for sign-in.
2. Connecting Google on the Plugins page (optional)
- What we ask for: one permission,
drive.file. It lets Model Stack create a Google Doc, and open only files it created itself. We don't ask for any access to your Gmail, your other Drive files, your calendar or your contacts. When a task writes an email, we only give you an "Open in Gmail" button that opens a new message on Google's own site with the text filled in; Google sees nothing until you press it, and we never read or send your email. - What it does with it: only when you ask a task to write a document. We then save the finished text as a new Google Doc in your account. We never read or send your email.
- What we keep: a refresh token (so we can do that later without asking you again), the permissions you granted, and the time you connected. The text of a document goes to Google to be saved, and a copy of your task and answer is kept the same way as any other task (see "What we collect").
- Disconnecting: press Disconnect on the Plugins page. We revoke the token with Google and delete our copy. You can also remove access at myaccount.google.com/permissions. Docs already created stay in your account until you delete them.
Who sees it, for both: only the Model Stack service that runs your account, and the AI companies that write your task's text, as described above. We don't share Google data with advertisers or data brokers, we don't sell it, and we don't use it to train AI models. To have your account and the Google-provided details deleted, email modelstack44@gmail.com.
Model Stack's use and transfer of information received from Google APIs adheres to the Google API Services User Data Policy, including the Limited Use requirements.
How long we keep it
Password-reset codes expire after 15 minutes. We keep your account and billing records while your account is open, and billing records for as long as the law requires afterwards. Things you save, such as agents, automations and memory, stay until you delete them. You can delete each of them yourself in the app.
Your choices
- Turn memory off, or clear what it remembers, on the Memory page.
- Delete agents, automations and scheduled tasks from the sidebar.
- Disconnect any linked account or signed-in website from Plugins. For a website this erases the saved sign-in.
- Ask us to see, correct or delete your account and the data tied to it by writing to modelstack44@gmail.com. We'll act on it within a reasonable time, though some billing records must be kept by law.
Security
There is a plainer page on this at Data and security. In short: our AI provider keys live only on our server and are never shown to customers. Passwords are hashed, sign-in attempts are rate-limited, and the site is served over HTTPS. No service is perfectly secure, so please use a strong, unique password.
Cookies and browser storage
- A sign-in cookie keeps you signed in. It can't be read by page scripts, and it is used for nothing else.
- Your browser's local storage holds your recent chat history and a few display preferences on your own device. We don't read it on our servers, and clearing your browser data removes it.
- Voice input. The microphone button uses your browser's built-in speech recognition. Depending on your browser, it may send what you say to the browser's provider to turn it into text. Model Stack never receives or stores your audio, only the text you choose to send.
- We don't use advertising or cross-site tracking cookies.
Your rights
You can ask us to give you a copy of the information we hold about you, to correct it, or to delete it. Write to modelstack44@gmail.com from the address on your account. We'll act on it within a reasonable time. Some billing records must be kept by law, and we'll tell you if that applies. If you are in a place with data protection laws, such as the EU, the UK or California, you have the rights those laws give you, and we'll honor them.
Who this is for
Model Stack is a business service. It isn't directed at children under 13, and we don't knowingly collect their information.
Changes to this policy
If we change this policy in a meaningful way, we'll update the date at the top and, where it matters, tell you in the app or by email.
Contact
Questions about privacy: modelstack44@gmail.com