Free tool
A sitemap either works or it does not, and the failures are quiet. Nothing warns you. The four we see most:
Nothing points at it. The file is fine and robots.txt never mentions it. Crawlers that do not guess /sitemap.xml never learn it exists.
It returns HTML with a 200 status. A misconfigured server answers the sitemap path with a soft 404 page. The status code says success, so retries do not happen, and the crawler reads an empty sitemap forever.
Cross-host URLs. The sitemap lives on one host and lists URLs on another. Those entries are discarded, silently, in full.
An index pointing at dead children. The index parses, declares twelve children, and four of them 404. You get two thirds of the pages you think you are submitting.
A sitemap is plumbing. It gets your pages considered for indexing; it does not make them rank, and it certainly does not make an AI engine cite you. Sites with immaculate sitemaps go uncited every day, because citation is decided by what other people wrote about your category, not by how tidily you enumerated your own URLs.
It is on this list because it is a ten-minute check that occasionally turns up something genuinely broken, not because it is the lever. If your sitemap is clean and you are still invisible in AI answers, the problem was never here, and the visibility checker is where you find out where it actually is.
Next
Whether the file is reachable at a path a crawler would try, whether it parses as XML, whether robots.txt points at it, how many URLs it really submits, whether any of those URLs sit on a different host, and whether lastmod dates exist. It follows a sitemap index one level down and reads the first few children.
Because it is how most crawlers find the file. /sitemap.xml is a convention, not a rule, and a crawler that does not guess the path has no other way of knowing the file exists. A perfectly good sitemap nobody points at is the single most common finding this tool produces, and the fix is one line.
Indirectly and modestly. Answer engines mostly retrieve through search indexes rather than crawling your sitemap directly, so a sitemap is not what gets you cited. What it does is make sure your pages are in the index that gets searched in the first place. Necessary plumbing, not a growth lever, and anyone selling it as the latter is selling something.
Yes, on purpose. This is a free tool fetching a stranger’s server, so it opens at most five children of an index. That is enough to confirm the index resolves to real URL sets. The count it reports covers only what it read, and it says so in the result rather than implying it counted everything.
A sitemap may only list URLs on the same host it is served from. Crawlers discard cross-host entries outright, so if your sitemap at example.com lists www.example.com URLs, or vice versa, those pages are not being submitted at all. This one bites sites that changed canonical hosts and left the generator configured for the old one.
Search crawlers use it to prioritise what to re-read, and they discount it when it is obviously wrong. A CMS that stamps today’s date on every page in every build is worse than no lastmod at all, because it trains the crawler to ignore the field. If yours updates honestly, it helps your genuinely-changed pages get recrawled sooner.