Privacy
urlhub uses no cookies, analytics tools or ads. We only count anonymous totals of views and clicks, with nothing about people (see below). Below we describe which data arises when you use it and what happens to it.
Who is responsible
The data controller is Adam Tomaś. For anything about your data, write to kontakt@doner.cloud.
The links you enter
The links you enter go to our server, which fetches the pages they point to or asks the services they belong to (YouTube, Vimeo, Spotify, SoundCloud, TikTok, X, Wikipedia, GitHub). Those pages and services see our server's address, not yours. Results are kept in the server's memory for a short time (up to 30 minutes, videos up to an hour) and are gone when it restarts. If you paste or load from a file a whole text (such as an email), it goes to the server only to find the links in it; the text itself is not stored anywhere. From rich text (such as an email or a document with links behind words) your browser takes the links out itself, and only they go into the field and on to the server. We do this to show the preview you ask for (legitimate interest, Art. 6(1)(f) GDPR). Entering links is voluntary, but without them there is nothing to show. Do not enter links that themselves hold data you do not want to pass on, such as addresses with a password or token.
Preview pictures
The pictures in the tiles and the table are fetched by our server – from the servers of the sites the links lead to – made smaller and only then shown to your browser. Those servers therefore see our server's address, not yours, and cannot set cookies on you. The smaller pictures are kept briefly in the server's memory (gone at a restart) and by your browser for up to a day. Links straight to video files are shown without a preview, since the film would play from another server. For a short link (/c/…) we make one preview picture for messengers: up to four such smaller pictures of its links, the number of links and the name urlhub.cloud. It is stored on our server for as long as the short link itself (it goes with it, and leaves the backups 14 days later), with no IP address and nothing about who made the link. We make the same picture for a list kept in the page address (?urls=…) when a person opens that page or a messenger fetches it. It is kept for 30 days from the last time it was fetched, under a name worked out from the links, with no IP address and without the list itself, which stays in the server's memory only until the picture is made.
The page address and short links
The list of links, the view, the advanced option and the language are part of the page address, so it can be shared. A short link (/c/…) is made with a button – or by the page itself when the list is too long to fit in the address – and stores only these four things on our server, with no IP address and no account; anyone who has it sees the list. A short link nobody opened for 90 days is deleted, and it leaves the daily backups 14 days later.
Published lists (pages with a title)
In the “Share” dialog you can lock a set’s content (the “Lock the content” button), which makes a locked set (/l/…). Our server then stores the title and description you give, the links, the view, the language, the date, and the pages' titles, descriptions and picture addresses as they were at that moment – with no IP address and no account. We do this at your request, to provide that page (performance of the service, art. 6(1)(b) GDPR). Anyone who has the link sees the page; it is kept out of search engines. A published page cannot be changed or deleted on the site, so do not put personal data in its title or description. We review published pages and delete those that are illegal or misleading (for example, leading to scam sites). A page nobody opened for a year is deleted, and it leaves the daily backups 14 days later. To have your page deleted or to report illegal content, write to kontakt@doner.cloud (the “Report abuse” link is on every published page).
Kits with a panel (pages to change)
In Share you can create a set you can change (/k/…): a page whose links you change later while its link and QR code stay the same; in the panel you can also copy it, which makes a new set with the same links. Our server then keeps its title and description, links, language, the dates it was made and changed, the titles, descriptions and picture addresses of the pages at the time of saving, its 20 previous versions (date, title, links) and a hash (SHA-256) of its panel key – no IP address and no account. The key itself is only in the panel's link you get once; we do not keep it. For the author we count each day how often the page was opened, how often through its QR code and how often each of its links was clicked – without cookies, IP addresses or identifiers; the same address within 10 minutes counts once (the address is only briefly in the server's memory for this), robots are not counted. In the panel you can also sign the set with your site's name and domain: we then keep the name, the domain, a random token, the dates of the proof and of the checks and the address of the site's icon. To prove the domain our server fetches its home page or the file /.well-known/urlhub.txt and reads its DNS records, when you save it, when the set changes and every 30 days. The signature shows on the set's page only once proved; you remove it in the panel. Once the domain is proved you can add a logo: we keep it as a WebP picture made on our server, without what the file stored besides the picture (such as where a photo was taken); it shows only with a proved signature, you remove it in the panel, and it goes with the set. We also keep the look you choose in the panel (theme, colour, face, cover, layout). We also keep the advertising labels you pick for links, with the brand's name, and your notes on links and section names. We do this at your request (Art. 6(1)(b) GDPR). Anyone with the page's link sees it; it is kept out of search engines. We review sets and delete those that are unlawful or misleading. You delete a set in its panel; a set nobody opens for a year is deleted with its statistics (counts older than 400 days go sooner), and it leaves the daily backups 14 days later.
Reports, appeals and removed sets
When you report a set or write to kontakt@doner.cloud, we get your email address, your name (if you give it) and your message. We use them to confirm receipt of the report, decide on it, answer and handle an appeal: a legal obligation under the Digital Services Act (DSA, art. 16 and 17; art. 6(1)(c) GDPR). Mail runs on OVHcloud servers in the European Union. We keep messages for 6 months after the decision, the time for an appeal, unless the case goes to an authority or a court. When we remove a set, we keep its address, the reason and the date for 6 months, so that its page can say why it is gone and how to appeal. Nothing about the author or the reporter is kept with it.
Lists featured on the home page
When you publish a list you can tick “Ask for it to be featured”. We then also keep the page's id, its language, the date you asked and our decision – no IP address. We go through the requests by hand and show the lists we pick on the home page under “Sets from our users” and on the page of all such sets (title, description, number of links, site names and up to three pictures from the links). We do this at your request (Art. 6(1)(b) GDPR). We may not show a list, or take it down, without giving a reason. A featured page is not deleted after a year while it is featured. We check the links of featured lists once a day; if none of a list's links works three days in a row, we delete its page. To take your list off the home page, write to kontakt@doner.cloud.
AI assistants (Claude, ChatGPT and others)
If you use urlhub through an AI assistant – as a connector (an MCP server) in Claude, ChatGPT or another assistant, or through a GPT that uses our API – the assistant sends our server only what the task needs: the links for a set or a preview (with the title and description of a page, if it publishes one), the address of a set to read, or the language of the featured sets. Your conversation with the assistant does not reach us – the assistant's provider processes it under its own terms. We handle the links just like links entered on the site (see above). A set made by an assistant is an ordinary short link (/c/…) or, when the assistant gives it a title, a page published as on the site (/l/…, see above: public, kept a year from the last visit, open to reports); such pages are marked for our review. The requests come from the provider's servers, so we see its IP address, not yours; we hold it briefly in the server's memory, only for the limits. Each use of an MCP tool is counted in the statistics as a total per day, with nothing about who asked. We do this to do what you ask the assistant for (legitimate interest, Art. 6(1)(f) GDPR).
Statistics without cookies
To learn which pages and sets are useful, our server counts totals for each day: how often a page was opened (each featured set on its own, people's published pages all together), where a visit came from – only as a kind (a search engine, ChatGPT or another AI assistant, a messenger, Hacker News, Product Hunt, AlternativeTo, Reddit, another site, directly), without the address of the page you came from – how often a link in a featured set was clicked (not which one), a QR code opened or downloaded, a link to a list or a set as text copied, “Make one like it” clicked under one of the home page's three examples (which one), campaign parameters added to links (not what they say), a page published or a short link made, how often programs and AI assistants read sets as text, and how often AI assistants used each of urlhub's tools (MCP). Your browser tells our server about moving between pages and about clicks. Until 11 November 2026 we also counted how often the note about a QR code whose list you can change after printing was shown and clicked (“I want to use it”); those counts no longer grow. We store no IP address, no cookie, no identifier and nothing in your browser, and we do not recognise the same person on a later visit; the IP address is only held briefly in the server's memory to keep one person from inflating the counters. Robots and messengers' link previews are not counted. We also count how many people clicked the invitation “Make your own set” on someone's set and how many of them made a short link, a page or a QR code in the same visit – the mark of where they came from is only in the address and then in the open tab's memory, not in a cookie. For each published page (/l/…) we also count one total of how often it was opened, with no dates and nothing about who opened it; opening the same page again from the same address within 10 minutes does not count. From 10 openings the number is shown on that page to anyone who opens it, and it goes with the page. The totals are kept for 13 months, the note's counters for a year. We do this to improve the service (legitimate interest, Art. 6(1)(f) GDPR).
Cookies and browser storage
urlhub stores no cookies. Recent queries go to your browser's storage (localStorage) only at your request, after you turn on “Remember recent sets on this device”. They are never sent anywhere, and turning the switch off deletes them. Without the switch, Recent shows only this visit's queries: they are kept in the open tab's memory and are gone when it is closed or reloaded. At your request (“Remember the panel on this device”) we keep a list “My sets” in your browser's storage: the title, the panel's address with its key and the date. It is never sent anywhere; “Forget” deletes it. Anyone who can use this browser can open a panel kept there. The light or dark theme picked with the switch in the header is also remembered in your browser, only once you click it.
Server logs
The server logs a shortened IP address (without its last part, so it does not point to a particular device), the date and time and the kind of request, without the links you entered. They are used only to keep the site running and secure (legitimate interest, Art. 6(1)(f) GDPR) and are overwritten by newer ones once they fill 30 MB, usually after a few weeks. The full IP address is held in the server's memory only, just to limit the number of requests and publications from one address, and is deleted when that limit's time is up – usually within a quarter of an hour, a day for the daily limit on publications – at the latest when the server restarts. If your browser finds something on our pages that breaks their security rules (Content Security Policy), it sends us a report with the address of the page and of that thing; we keep it in the technical logs, without an IP address. The web server in front (Apache) keeps no log of visits; only on technical errors its error log may hold an IP address and the page's address, and it is deleted after 30 days.
Where the data is and who receives it
The server runs in an OVHcloud data centre in France (European Union), on our behalf. Links also go to the services we ask about them; some of them (such as YouTube and GitHub) are based in the USA, but they do not get your IP address. We do not sell data, do not profile you and make no automated decisions about you.
Cloudflare
Traffic to the site goes through Cloudflare's network, which protects it from attacks and speeds it up (legitimate interest, Art. 6(1)(f) GDPR). Cloudflare processes the IP address and technical request data as described in its privacy policy. The company is based in the USA and transfers data under the EU-US Data Privacy Framework.
The “Buy me a coffee” button
The “Buy me a coffee” button in the footer is a plain link to buycoffee.to: the site does not connect to buycoffee.to until you click it, and after the click buycoffee.to processes your data under its own terms.
Your rights
You have the right to access, rectify and erase your data, to restrict its processing and to object to it. You can also lodge a complaint with the supervisory authority: the President of the Personal Data Protection Office of Poland, UODO (uodo.gov.pl).