According to the HTML5 specification (section 8.1.2.3. Attributes): "Attribute names must consist of one character other than the space characters, U+003E GREATER-THAN SIGN (>), and U+002F SOLIDUS (/), followed by zero or more characters other than the space characters, U+003E GREATER-THAN SIGN (>), U+002F SOLIDUS (/), and U+003D EQUALS SIGN (=)." Currently we only check for the presence of '/' in attribute in HTMLTokenizer::addAttribute.
Whoever handles this bug, changes in the assertion should be matched with changes in the parser code to avoid hitting the assertion. Check bug 13942 for the details.
The spec has changed not to include the text you quote, but I believe you're quoting an authoring requirement and not a user agent conformance requirement. If that's not accurate, please re-open this bug with a reference to the user agent conformance requirement you believe we violate.