$ curl cheat.sh/

# xmllint
# XML parser and linter that supports XPath, a syntax for navigating XML trees.
# More information: <https://manned.org/xmllint>.

# Return all nodes (tags) named "foo":
xmllint --xpath "//foo" source_file.xml

# Return the contents of the first node named "foo" as a string:
xmllint --xpath "string(//foo)" source_file.xml

# Return the href attribute of the second anchor element in an HTML file:
xmllint --html --xpath "string(//a[2]/@href)" webpage.xhtml

# Return human-readable (indented) XML from file:
xmllint --format source_file.xml

# Check that an XML file meets the requirements of its DOCTYPE declaration:
xmllint --valid source_file.xml

# Validate XML against DTD schema hosted online:
xmllint --dtdvalid URL source_file.xml

$

Follow @igor_chubin cheat.sh tldr