Add a website checker to your site.

Yes. Pick Is it down?, the port checker or the website file checker and a theme above, and copy the snippet: an iframe with the checker and a link to the full tool. It's free, needs no account or key, and your visitors' checks count against their own limit, not yours. The snippet loads lazily and sets no cookies on your page.

It requests the website's home page from Washington, D.C., United States, Frankfurt, Germany and Singapore at the same time, the way a browser would, and notes the status code and how long the server took to answer. It also tries the site from your own browser. Up everywhere but not for you means the problem is on your side; down everywhere means it's down for everyone.

Try another network, such as your phone's mobile data. Flush your computer's DNS cache, or restart the router. Turn off any VPN, proxy or ad blocker and try again. If the site moved to a new server recently, your DNS may still have the old address cached for a while.

"Doesn't point to any server" means the domain has no DNS records for its website, often because they were deleted or the domain expired. "Refused" means the server is reachable but no website is running. "No answer" means the server or a firewall ignored the request. A certificate error means browsers will refuse the site until the certificate is renewed. Status codes of 500 and above mean the server is running but the website is failing.

No. The server is up and answering, but the home page is refused or missing. Visitors see an error page, so it's worth fixing, but the site itself isn't down.

80 (HTTP), 443 (HTTPS), 25 (SMTP), 465 (SMTPS), 587 (Submission), 993 (IMAPS), 995 (POP3S), 21 (FTP), 22 (SSH / SFTP), 53 (DNS), 8080 (HTTP alternate), 3306 (MySQL). Only these, and only on the address a domain or host name points to. IP addresses can't be entered, so the checker can't be used to scan networks.

Open means a service accepted the connection. Closed means the server is reachable but nothing is listening on that port. No answer means nothing came back in time, usually because a firewall drops connections to the port, which is what you want for ports you don't use.

A domain's email usually goes to other servers than its website, named in its MX records. The port checker lists them; check one of those hosts to see whether mail apps can connect on 465, 587, 993 and 995.

Cloud providers block outgoing connections on port 25, the port mail servers use to pass email to each other, to stop spam being sent from their networks. Our servers run in the cloud, so the checker shows port 25 as Can't check rather than guessing. Port 587 on the same host usually shows whether the mail server is running.

Usually not. 3306 is MySQL, and a database open to the internet is a common target. Keep it closed to everyone except the servers that need it, or reach the database over SSH or a private network instead.

The files a site serves for crawlers, browsers and social apps: robots.txt, the sitemap it names (or /sitemap.xml), the home page's Open Graph and Twitter card tags with the share image, the favicon and apple-touch-icon, and the web app manifest. It also spots a noindex tag on the home page. Each finding says what's wrong and how to fix it.

It's a plain text file at /robots.txt that tells crawlers which parts of a site they may fetch, and where the sitemap is. A single line, Disallow: / under User-agent: *, tells every search engine to stay away, which is a common leftover from a staging site. The checker reads the rules the way Google does and says whether Google, Bing and other crawlers may reach the home page.

Not always. robots.txt stops crawling, not indexing, so a blocked page that other sites link to can still show up in results, without a description. To keep a page out of search, let it be crawled and add a noindex robots meta tag instead.

It's your choice. GPTBot, ClaudeBot, Google-Extended, CCBot and others each have a user agent you can disallow in robots.txt to opt out of AI training or AI search. The website file checker lists which ones a site blocks, and doesn't count either choice as a problem.

Small sites with good internal links do fine without one, but a sitemap helps search engines find new and deep pages sooner, and tells them when pages change. Name it in robots.txt with a Sitemap line. One sitemap file can list up to 50,000 URLs and be up to 50 MB; bigger sites split it into several files and list them in a sitemap index.

Apps like Facebook, LinkedIn, Slack and iMessage read the og:image tag in the page's head. If it's missing, isn't a full https:// address, doesn't load, is an SVG or is very small, they show no picture or a small thumbnail. Use a 1200×630 PNG or JPEG under 5 MB, and add twitter:card summary_large_image for a large card on X.

Google shows a favicon next to search results when it's square and at least 48×48 pixels. A 48×48 or larger PNG, or an SVG, linked with rel="icon" covers browsers and Google, and a 180×180 apple-touch-icon covers iPhone and iPad home screens. Browsers also ask for /favicon.ico when a page links no icon.

Without an account, 30 Is it down? checks, 20 port checks and 30 website file checks an hour per IP address. With a free account, the REST API, MCP server and CLI allow 500, 300 and 500 a day.

Yes. The Lemon Domains MCP server has check_website_status, check_ports and check_site_files tools, the REST API serves them at /api/v1/is-it-down, /api/v1/ports and /api/v1/site-files, and the lemon CLI runs lemon down example.com, lemon ports example.com and lemon site-files example.com. Agents without an account can fetch lemon.domains/tools/is-it-down/example.com.md, or .json.