Last updated: 6 November 2026
Privacy Policy
This page describes what the current Everything Hub website and its connected features do with information. It is a practical description of the implementation, not a promise about the separate services linked or embedded here. Their practices and the site’s features can change.
Who provides the site
Everything Hub is a static website hosted on GitHub Pages. Supabase provides the account, database and server-side function services used by the site. AI chat uses KindleHub's API at kindlehub-api.arancool3000.workers.dev when `KINDLEHUB_API_KEY` is configured; otherwise it can use Google Gemini when `GEMINI_API_KEY` is configured. These third-party providers process requests under their own terms and privacy notices. Everything Hub does not control KindleHub's API operation, data retention or availability.
Information used by the site
Accounts and sign-in
Depending on the sign-in method, account information can include a username or email address, authentication credentials, a display name or chat nickname, and selected subjects. Supabase Auth handles passwords and sign-in sessions; passwords are not included in the saved progress record. Username-only accounts use an internal sign-in identifier managed by the account service. Account email addresses are not shown to other people in chat. Username recovery codes are shown when generated; the recovery service stores hashes of those codes rather than the codes themselves.
Saved progress and device storage
When you sign in, the workspace can save lists, habits, tasks, notes, bookmarks, timetable and revision progress, rewards, profile-picture data and other account settings. The main progress record is stored in Supabase, is limited to 100,000 bytes, and is also cached in this browser so it can load and handle interrupted sync. Sync is not end-to-end encrypted; the service operator and service providers may be able to access stored data. Do not save information you would not want the service operator to see.
Some features also keep device-only data such as meal ideas, game scores, app pins and recent items, reading position and settings, and unsynced changes. The weekly Meal Planner stores the ideas you enter in this browser under a key for the signed-in account; they are not synced to Supabase. These local features may be associated with the signed-in account on that device. Signing out does not necessarily erase them. Clearing this site’s browser storage removes local data and identifiers, but does not delete the account or cloud data; unsynced changes may be lost.
Community and shared content
If you choose to publish a flashcard set, it is visible to other signed-in users. Community chat messages and the nickname used with them are visible to signed-in users. The current chat code removes messages older than 24 hours during later message activity and caps the stored message list at 1,000; because cleanup runs when messages are sent, a message may remain longer if no later message triggers cleanup. You can delete your own community message in the app. Reports and moderation-related records may also be processed.
Live-game rooms contain the host’s questions, players’ nicknames and answers, and any room-chat messages. Rooms expire after about two hours and are removed during later room creation; related room data is deleted with the room. Expired data may remain until that cleanup runs. Use a nickname that does not identify you, and do not post contact details or other private information.
Signed-in activity counts
The site’s first-party status counter records activity only while an account is signed in. It uses a random browser identifier saved in local storage, a random per-tab session identifier, the signed-in account identifier and timestamps to count page views and recent presence. The private status page is excluded from page-view counts. Its counting function does not store page paths, page text, questions, IP addresses or email-address fields. Supabase may still process ordinary request metadata, such as a source IP address, while serving requests under its own service practices. The browser identifier helps estimate distinct browsers; it is not a reliable count of people and changes if browser storage is cleared.
Session rows older than one day are removed when a later activity ping performs cleanup. Unique-browser identifiers and aggregate totals have no automatic expiry in the current database code. Historical anonymous totals from before anonymous tracking was disabled remain part of the displayed historical counts; new anonymous page views are not collected.
AI chat
If you choose to send a question, only the current message (up to 4,000 characters) is sent from the browser to the Everything Hub Supabase function; earlier chat messages are not sent. When `KINDLEHUB_API_KEY` is configured, the function sends that message server-side to KindleHub's OpenAI-compatible API at `kindlehub-api.arancool3000.workers.dev`. Otherwise, it uses Google Gemini if `GEMINI_API_KEY` is configured. Provider keys remain server-side. The displayed conversation stays in browser memory while the page is open; clearing it only clears that display. Supabase stores the account identifier and the latest short-term request window start/count for abuse prevention, not the prompt or answer. There is no per-account daily site cap; the server permits up to 12 attempts per account in a fixed 60-second window. The limiter keeps one row per account and the row is removed if that Auth account is deleted. Failed or cancelled requests may count towards the short-term limit. KindleHub says API calls share the key owner's daily AI allowance with their KindleHub app; its current documented Free, Plus, Pro and Max tiers have different daily message, per-key rate and output limits, and free requests may be refused first when shared capacity is full. This site's users share the configured key's allowance. KindleHub describes its Developer API as invite-only and prohibits sharing keys. KindleHub's terms and privacy practices govern its processing; its retention and availability are outside our control, and paid plans may incur charges. If Gemini is used instead, Google's terms, model quotas and any project billing settings apply and may incur charges if billing is enabled. Do not submit personal or sensitive details.
Browser storage and third-party services
The site uses browser local storage and session storage for sign-in state, saved or cached progress, the activity identifier, settings and device-only feature data. The core app does not load a third-party analytics SDK. The signed-in counter described above is first-party and uses a local-storage identifier rather than an advertising cookie.
Some features contact external services when you view or use them. The landing-page Launch Orbit, Better Launch and CodeHype badges load images from those providers. Video thumbnails and embedded lessons use YouTube services, including YouTube’s privacy-enhanced player domain. Optional book scans use Internet Archive. Games may load from Geometry Dash, Two Player Games, Classroom Resources, jsDelivr or the independent game collection. Links and saved bookmarks can open other websites. These services receive their own requests and may receive information such as your IP address, browser/device information and a referrer, and may use cookies or similar technologies. Their availability, content and data practices are controlled by those providers, not Everything Hub. Review the relevant provider’s privacy information before using an embed or external link.
Retention and account requests
Cloud progress and account information remain in the connected service until changed or removed by the service operator. The site does not currently provide a self-service account deletion control. AI rate-limit data keeps one latest-window row per account rather than a daily history; it is removed when the associated Auth account is deleted. Provider logs, backups and operational records may have separate retention rules set by the providers. Local browser data remains until it is overwritten or cleared in your browser.
Questions or deletion requests: contact the maintainer through the project’s GitHub page. GitHub issues are public, so do not post your username, email address, recovery code, password, AI prompt or other private account information there. No private support form or self-service deletion flow is currently provided on this site.
Changes to this policy
This policy may be updated when the site’s data practices or connected services change. The date at the top indicates the latest revision.