Web fetcher

About BrightPresenceSiteCheck

Last updated: August 2026.

In plain English If you found this page in your server logs, one of my jobs asked your server for something. Usually it is a single request for your home page, to see whether the site loads and whether its security certificate is valid. One of my jobs keeps a few short quotations from that page, and a second fetcher looks for a published contact address on a handful of pages. No forms, no logins, nothing behind a password. Every job I run reads your robots.txt first and stops when it says no. If you would rather I did not come back at all, add two lines to your robots.txt or email me the domain.

What it is

Bright Presence Digital LLC builds and maintains websites for schools, municipalities and small businesses. Before I ever write to an organization, I check one thing about the address I hold for them: does it load, and does its HTTPS certificate actually cover the domain? A certificate that has expired or was issued for a different name shows every visitor a full-page browser warning, and most owners find out from a parent or a customer rather than from a monitoring tool.

BrightPresenceSiteCheck is the fetcher that answers that question. It identifies itself in every request:

Mozilla/5.0 (compatible; BrightPresenceSiteCheck/1.0; +https://brightpresencedigital.com/sitecheck.html)

Exactly what it requests

/robots.txt, first, once per hostname.
The one page of your site I hold a link to (usually the home page) once, following redirects.
A TLS handshake on port 443, to read the certificate's expiry date and the names it covers.

That is the whole list. It does not submit forms, follow links deeper into the site, attempt any login, or request files it was not pointed at.

It does keep some of your wording, and I would rather say so plainly than let you find it. The site and certificate check reads the HTML only to work out what platform the site is built on, and keeps that one answer. A second pass compares a fingerprint of the page's visible text against the last visit, and when the text has changed it keeps up to five short passages, each between 15 and 300 characters, copied exactly as your page has them, with a one-line note about each. Email addresses and phone numbers are removed before anything is stored.

Those passages are read out of the page by a language model running on my own hardware, in my own office. Nothing from your page is sent to any outside service. No third party sees it, and it is not used to train anything.

How often

At most once per hostname per day, and in practice far less. A given site is typically looked at once every several weeks. Organizations that share a hostname are checked once between them, not once each.

Three jobs use this identity. One checks whether a site loads and whether its certificate is valid. One re-reads the page to see whether its content has changed since I last looked. One reads the page once to work out what platform the site is built on. All three request a single page, all three read your robots.txt first, and all three stop if it tells them to.

One other fetcher of mine may reach you, and it identifies itself separately as BrightPresenceContactFinder/1.0 with a link back to this page. Before I write to an organization I look for a published contact address, so it requests up to eight pages on your site: the home page, /contact, /contact-us, /about, /about-us, /admissions, /staff and /faculty, pausing between each one. It reads your robots.txt first and stops when told to, it keeps only the email addresses your site already publishes, and it keeps nothing else from the page. The two lines below stop it as well.

How to block it

Add this to your robots.txt and the fetcher will skip your site entirely on its next pass:

User-agent: BrightPresenceSiteCheck
Disallow: /

A blanket User-agent: * disallow is honoured too. A Crawl-delay directive is read and respected. If your robots.txt cannot be reached, I treat that as permission unchanged rather than as a refusal, so a temporary outage on your side does not silently change what I do.

You can also block the user-agent string at your firewall or WAF. Match on the name BrightPresenceSiteCheck rather than on a whole line, and add BrightPresenceContactFinder if you want the contact lookup stopped that way too. I will not work around a block, change my user-agent to evade one, or come back from a different address. The only time I ask twice is when your server replies with a rate limit, and then it is the same name from the same address, after the delay your server asked for.

What I keep

For each hostname: whether it responded, the HTTP status, the platform the site appears to be built on, the certificate's remaining days, the name it was issued to and whether it covers your domain, any error, and the time I looked. I set and read no cookies, and I keep no copy of your HTML.

I also keep some of your page's text, and here is exactly what that means. On the pass that checks whether your site has changed I store a one-way fingerprint of the visible text and how long that text was. When it has changed, up to five short passages are kept word for word, 15 to 300 characters each, with a one-line note about why each stood out. Because they are copied exactly as your page has them, a passage can include a staff member's name or job title where your page prints one. Email addresses and phone numbers are stripped before anything is stored.

That material is assembled into a short summary which is shown to the person who makes my sales calls, so that a call to your office is informed rather than cold. Short passages like these may also appear on a sample redesign of your home page that I build to show what a rebuild could look like. Those samples sit at an unlisted address, nothing links to them, and they are not indexed.

Nothing here expires on its own. If you want it gone, see below, and it goes.

Questions, or want your domain removed

Email scott@brightpresencedigital.com with the domain. A person reads that inbox. Your domain goes on a permanent do-not-fetch list that every one of my jobs checks before it even looks up your address, so you stop seeing requests from me entirely, including the one for your robots.txt. Everything I already hold from your site is deleted: the quotations, the summary, the fingerprints and any sample page built from your content. That list is not reversed by anything else changing on my side.