Privacy Policy

Last updated: October 20, 2026

This privacy policy describes how Business Things and Whatnot, LLC (“we,” “us,” or “our”) collects, uses, and shares information when you visit thomasjfrank.com (the “Site”).

Information you provide directly: When you subscribe to our email newsletter, we collect your email address. This information is provided voluntarily when you complete a signup form on our Site. If you use the documentation assistant, your questions, the answers, and any rating or feedback you give are handled as described under Documentation Assistant.

Information collected automatically: When you visit our Site, we automatically collect certain information about your device and your visit, including your IP address, browser type, operating system, referring URLs, and information about how you interact with the Site. We collect this information using cookies and similar technologies.

We use the information we collect to operate and improve our Site, send you email newsletters and updates you have requested, analyze Site traffic and usage patterns, protect against spam and abuse, deliver and optimize content, and answer questions you submit to the documentation assistant.

If you are located in the European Economic Area (EEA), United Kingdom (UK), or Switzerland, we rely on the following legal bases to process your personal information:

Consent: We process your email address for newsletter communications only with your consent. You can withdraw consent at any time by unsubscribing.

Legitimate interests: We process analytics and technical data to understand how visitors use our Site, improve our content, and ensure the security of our services.

Contract performance: If you purchase products from us, we process your information as necessary to fulfill your order.

Documentation assistant: When you submit a question, we process it to generate an answer, to limit abuse, and to keep a log we use to improve our documentation and the assistant. We rely on our legitimate interest in answering questions about our documentation, improving it, and protecting the Site. Using the assistant is optional.

We use the following third-party services that may collect information about you:

Kit (formerly ConvertKit): We use Kit to manage our email newsletter. When you subscribe, your email address is stored by Kit. Kit may also collect information about how you interact with our emails. You can review Kit’s privacy policy at kit.com/privacy.

Google Analytics 4: We use Google Analytics to understand how visitors use our Site. Google Analytics collects information such as how often you visit, what pages you view, and what other sites you visited before coming to our Site. Google Analytics uses cookies to collect this information. You can learn more about how Google uses data at google.com/policies/privacy/partners.

Cloudflare: We use Cloudflare for content delivery, performance optimization, and security. Cloudflare may collect technical information such as your IP address to provide these services. You can review Cloudflare’s privacy policy at cloudflare.com/privacypolicy.

YouTube: Our Site contains embedded YouTube videos. These videos use a privacy-enhanced embed that only loads when you click to play. Once loaded, YouTube may collect information about your viewing activity in accordance with Google’s privacy policy.

Lemon Squeezy: If you purchase products from us, the checkout process is handled by Lemon Squeezy on their hosted pages (shop.thomasjfrank.com). Lemon Squeezy collects payment and order information as described in their privacy policy at lemonsqueezy.com/privacy.

Circle: We use Circle to host our community platform at community.thomasjfrank.com. When you join our community, Circle collects your account information and activity within the community. You can review Circle’s privacy policy at circle.so/privacy.

OpenRouter: The documentation assistant sends questions to OpenRouter, which routes them to a model provider. When the assistant searches or fetches a page, that query goes to a search or fetch service. The sites a search is allowed to use, and what is kept, are described under Documentation Assistant. You can review OpenRouter’s privacy policy at openrouter.ai/privacy.

Notion: We store a log of documentation assistant conversations in a private Notion database (see Documentation Assistant). Notion hosts that data on our behalf. You can review Notion’s privacy policy at notion.com/notion/privacy.

On Ultimate Brain and Creator’s Companion documentation pages, we offer an assistant that answers questions about those docs.

What you submit. The question you type, and up to eight recent messages from that same conversation. Each of those is limited to 2,000 characters. You do not need an account. We do not send your name, email address, or an account id with the question.

Where it goes. Our server receives the question, selects relevant pages from our documentation, and sends a prompt to OpenRouter (openrouter.ai). The prompt includes your question, those recent messages, excerpts from the documentation pages, and our instructions to the model. OpenRouter then routes the prompt to a third-party model provider. The provider can be different from one question to the next, and it may process the prompt outside the United States.

We configure this OpenRouter project so a question is routed only to providers that do not train on prompts and that offer zero data retention. Each request also tells OpenRouter to require zero data retention. Zero data retention means the provider uses the prompt to produce the answer and does not keep the prompt or the answer afterward. A provider may still hold a prompt briefly in memory so a repeated request costs less. That is not the same as saving it.

OpenRouter receives the prompt in order to route it and apply its filters. OpenRouter’s documentation says it does not store prompt or answer text unless the account turns on prompt logging, or turns on a setting that lets OpenRouter use prompts to improve its product. This project does not use those settings. OpenRouter does keep metadata about the request, such as the model name, token counts, and how long the request took. OpenRouter also says it may categorize a small sample of prompts without attaching that sample to an account, using a model that does not retain the text.

Before a prompt reaches a model provider, our server attempts to remove items such as email addresses, phone numbers, payment card numbers, and keys or passwords. OpenRouter’s filters then try to redact phone numbers, Social Security numbers, payment card numbers, IP addresses, and keys or passwords that remain. These filters do not remove names or street addresses, they can miss information, and they can also remove ordinary text. Do not rely on them.

OpenRouter also checks each request for common prompt-injection patterns, meaning text that tries to override the assistant’s instructions. Matching text is redacted before the prompt reaches a model provider.

Search is limited to four sites. If the documentation we already retrieved is not enough, the assistant may search, or fetch a page. It can only use these sites:

• thomasjfrank.com
• flylighter.com
• www.notion.com
• developers.notion.com

It cannot search the rest of the web. The OpenRouter project is set so a request cannot add other sites. A search query is based on your question and is sent to a search service OpenRouter is allowed to use for this project: a model provider’s own search, Exa, Parallel, Firecrawl, or Perplexity. A page fetch, when one runs, may be handled by a model provider, OpenRouter, Exa, Parallel, or Firecrawl. OpenRouter’s zero-data-retention rule for model providers does not cover search or fetch. Those services handle the query under their own policies.

What we keep. We keep a log of assistant conversations in a private Notion database so we can find gaps in our documentation and fix wrong answers. For each conversation, the log contains:

• each question, after our server has removed email addresses, phone numbers, keys, and similar items
• each answer, or the error if one could not be produced
• the documentation page you asked from and the pages the answer cited
• which product’s documentation you were in, the time you asked, and a random conversation id created by your browser
• technical details: response time, the model and provider used, token counts, cost, and OpenRouter’s request id
• any thumbs-up or thumbs-down rating you give, and any reason you type after a thumbs-down (cleaned the same way as a question, up to 1,000 characters)

The log does not include your IP address, your browser details, or an account id, and we do not use it to train our own models. This documentation may be accessed by people who work on this site, as well as Notion staff who may be able to access it as described in Notion’s own terms.

We do not log a question that is refused before it reaches the model—for example, one that is too long, sent while the assistant is turned off, or over a usage limit. Those requests receive a message pointing you to our support options.

In your browser, the conversation stays in that tab’s memory until you clear it or close the tab.

To limit abuse, we store a salted hash of your IP address with a count of recent questions and ratings. Those counts are deleted after the limit window, which is no longer than about a day. No question text is stored with that count.

If a request fails, we also keep the question, shortened to 180 characters, with the error and other technical details in server logs so we can fix the problem. We do not use those logs for advertising or to train a model.

How long we keep it, and deleting it. We keep the log for as long as it is useful for improving our documentation, and we delete older entries when they no longer serve that purpose. To have a conversation deleted, email us (see Contact Us) with roughly when you used the assistant and what you asked, and we will find and delete it.

We also send OpenRouter our site address and the name “Thomas Frank Docs Assistant” so the request can be attributed to this Site.

Do not send private information. Do not include anything private, confidential, or personal. That includes passwords, API keys, payment details, government ID numbers, health information, the contents of a private Notion workspace, and personal information about you or anyone else. Your question leaves our server. It is processed by OpenRouter and a model provider, a search query may be processed by a search service, and the question and answer are stored in our Notion log.

You can read the documentation without using the assistant.

Our Site contains affiliate links, including links to Amazon.com through the Amazon Associates program. When you click an affiliate link and make a purchase, we may earn a commission. These affiliate programs use cookies to track referrals. Amazon’s privacy practices are described in their privacy notice at amazon.com/privacy.

We use cookies and similar tracking technologies to collect information about your browsing activity. Cookies are small text files stored on your device. We use both session cookies (which expire when you close your browser) and persistent cookies (which remain until they expire or you delete them). You can control cookies through your browser settings. Note that disabling cookies may affect certain features of our Site.

A note about embedded YouTube videos: Our site does not immediately load embedded YouTube videos. You must click the placeholder image for an embedded video to load the player. Once you do, we use YouTube’s “no-cookie” embeds in an attempt to respect privacy as much as possible.

We do not sell your personal information to third parties.

Depending on your location, you may have certain rights regarding your personal information, including the right to access the personal information we hold about you, request correction of inaccurate information, request deletion of your information, opt out of certain data collection or processing, and withdraw consent where processing is based on consent. To exercise any of these rights, please contact us using the information below.

Email opt-out: You can unsubscribe from our email newsletter at any time by clicking the unsubscribe link at the bottom of any email.

We retain your personal information for as long as necessary to fulfill the purposes for which it was collected, including to satisfy legal, accounting, or reporting requirements. Email subscriber information is retained until you unsubscribe. Questions, answers, and feedback sent through the documentation assistant are kept as described in that section, including how to have them deleted.

Our Site is not directed to children under 18, and we do not knowingly collect personal information from children under 18. If you believe we have collected information from a child under 18, please contact us so we can delete it.

Service providers: We share your information with the third-party service providers listed in this policy as necessary to operate our Site and provide our services.

Business transfers: We may share or transfer your information in connection with, or during negotiations of, any merger, sale of company assets, financing, or acquisition of all or a portion of our business to another company.

Legal requirements: We may disclose your information where required to comply with applicable law, governmental requests, judicial proceedings, court orders, or legal processes.

Most web browsers include a Do-Not-Track (DNT) feature or setting. Because there is no uniform standard for recognizing or responding to DNT signals, we do not currently respond to DNT browser signals. If a standard for online tracking is adopted that we must follow in the future, we will update this policy accordingly.

We may update this privacy policy from time to time. We will notify you of any changes by posting the new policy on this page and updating the “Last updated” date.

If you have questions about this privacy policy or our data practices, please contact us at:

Business Things and Whatnot, LLC

500 Locust St. #121

Des Moines, IA 50309

Email: [email protected]