Paste an XML or HTML document, pick the matching parse mode, then type an XPath expression such as //div[@class='item'] or //book/title/text(). The document is parsed with DOMParser and the expression is evaluated using the browser's native document.evaluate, so results match the standard XPath 1.0 engine.
Matching nodes are listed with their type and serialized value; number, string and boolean results are shown directly. Syntax errors and parse errors are reported inline. Everything runs locally in your browser.