Trust, earned.
Everything on this page is something you can check. Where we can point at a test, a record, or a page that proves it, we do.
Last changed 5 October 2026. When something here changes, the date changes with it and the change is written on our status page.
What we never do
- We never touch your website. We read it from the outside, the way a visitor does, and we write down what came back. We hold no password, no login, no key to anything of yours, and we never ask for one.
- We never watch a site nobody has proved. Before a single reading is taken, the organization proves the website is theirs by adding one record to its domain's DNS. Nobody can put up a status page in your name, and nobody can point us at a website that is not theirs unless we have found a link to it, or the form embedded, on the website they proved.
- We never watch an outside address just because someone typed it. We watch the site that was proved, and any other address on that same domain a member asks for. Outside that domain, for example a donation page on another company's platform, we watch a page only if we have found a link to it, or the form embedded, on your own proved website, whether we found it ourselves (you then accept or decline it) or a member named it in Slack. Anything else is refused, with a sentence saying why. At most three pages outside your domain, each read every five minutes. Both rules, the link and the limit of three, are enforced in the database, not only in the software that talks to you.
- We never sell or share anything about you. There is no advertiser, no analytics that follows you, and not one cookie anywhere on this site.
What we hold, and for how long
Your email address (until you ask us to delete it), the addresses you watch (until you delete the organization; stopping a watch keeps the address with its history), what we saw when we looked (every reading for thirty days, then the readings are deleted and one line a day survives, being how many times we looked and how many were clean), your organization's name, and, if you installed the Slack app, the channel you chose and the address Slack gave us to post into it. The full table, with the reasons, is on the privacy page. It runs on Cloudflare, Supabase and Google Workspace, in the United States, each acting on our instructions and nothing else.
Who watches the watcher
A service that watches websites can itself fall over quietly, and its own status page would be the last to know. So an outside company, on computers we do not control, reads our runner every minute and writes to us when it stops answering. We tested it on 13 September 2026 by switching our own runner off on purpose: the email saying we were down arrived in seven minutes, and the one saying we were back followed the recovery. Our own status page showed the gap the whole time.
Our mistakes go on the record
Our status page carries a reading log, and the first entries in it are about us: an audit of our own work that found our own pages missing the protections we grade other sites on, a status page that briefly published the previous domain's dates under the new name, and a feed with the wrong weekday. Each entry says what was wrong and when it was fixed. A status page that hides its own findings is not worth trusting with yours.
How the pages you read from us are served
Every page we serve, this one included, carries the six browser protections we grade members on, is served only over HTTPS, and sets no cookie. Email from statuspost.org is signed by statuspost.org, and our forgery policy tells receiving mailboxes to quarantine anything pretending to be us. You can verify all of it from the outside, which is the point.
When alerts arrive
Alerts are live, and here is how they work, written down so you can hold us to it. We send an alert only when a page or form stops loading, your address can no longer be found, the padlock certificate is close to expiring or broken, or your domain's renewal is close; speed, email settings and browser protections never send one. The first sentence of every alert is written by code from the reading and states only what was measured. The paragraph after it says what is likely causing it and what to do; it is written by an AI model from those measured facts, is always marked as likely, and is allowed to say "we cannot tell." The facts sent to the model are the outage itself: the address we watch and what it answered. A question a member types to the Slack app is sent as typed, with those facts, so the answer can address it; nothing else about the person goes with it. The model providers are named on the privacy page like every other company that touches your data.
Who is responsible, and what we ask in return
V2C Inc, a Delaware nonprofit corporation, runs StatusPost; the volunteers who build it act for V2C. In return for a free service we ask three things, all in the terms: prove only what is yours, do not use us to burden anyone's site, and do not present our readings as something they are not. The limits that keep the service affordable for everyone are enforced by the service itself.
Reporting a security problem
Write to contact@statuspost.org. Our security contact is also published at /.well-known/security.txt, the standard place. We will answer, credit you if you want, and write down what we changed.
Leaving
Remove a check and we stop watching it. Ask us to delete your account and everything about you and your organization goes, confirmed by email. No form, no fee, no questions.