Break a URL into scheme, host, port, path, query and fragment, with the parts that commonly surprise people — encoding, userinfo, default ports — called out.
Waiting for a URL
Paste one and it is taken apart, with every parameter decoded.
Waiting for a URL
A scheme and a host are enough — the rest is optional.
URL Inspector runs entirely on your device — nothing is uploaded, and no URL is fetched.
You’re on the Free plan. Compare plans
A URL is read by a parser, not by eye, and the two disagree more often than people expect. This shows what a real parser makes of yours — the same one your browser and your server use.
The fragment is the part most worth understanding. Everything after `#` is never sent to the server: it is the browser's business alone. A token or a state value placed there is invisible in server logs, which is exactly why OAuth's implicit flow used it, and exactly why it is still in your browser history.
Percent-encoding is shown decoded beside the raw form, because `%2F` and `/` are different things. A `%2F` inside a path segment is a literal slash in that segment's value, not a separator, and a server that decodes before routing will split the path in a place you did not intend.
Userinfo — `https://user:pass@host` — is called out whenever it appears. It is valid URL syntax, it is deprecated for HTTP, browsers strip or warn about it, and it puts a password into every log and referrer along the way.
Nothing is fetched. The URL is parsed in your browser, so this cannot tell you whether the host exists or what it would return.
Never. Everything after `#` stays in the browser. That is why it was used to carry OAuth tokens, and why those tokens still end up in browser history and in anything that reads the address bar.
A `/` separates path segments; `%2F` is a literal slash inside one segment's value. Servers that decode before routing conflate them, which is a classic source of path-traversal and routing bugs.
Because it is the default for the scheme — 443 for https, 80 for http — and a URL parser normalises it away. A non-default port is kept, including `http://host:443`, which is unusual but meaningful.
It is valid syntax and a bad idea for HTTP. Browsers strip or warn on it, and the credentials travel into logs, referrers and shell history. Use an Authorization header instead.
No. Nothing is fetched. This tells you how the URL is structured, not whether anything answers at the other end.