html tag counter online
Paste an HTML snippet and get a count of every tag and attribute, plus the spots a visual read tends to miss: an image with no alt, a link with no text, and the heading list in the order it appears in the document.
The img tag has no alt attribute, so it lands straight in the missing-alt list; the link to /contato has an href but no visible text inside it, which is why it counts as empty.
br, hr and input have no </tag> in the original HTML and each one still adds 1 to the total count, exactly like the paragraph that has both an opening and a closing tag.
The class attribute shows up 4 times because the count is per tag occurrence, not per unique value; two divs and two spans, each carrying its own class, add up to 4.
No. The HTML is only parsed to build the element tree and count the structure, it is never inserted into the page DOM or rendered. As a result, any <script> in the markup is treated as text and never executed, which makes it safe to paste third-party HTML. All analysis happens locally in your browser.
It is not a syntax validator: it uses a lenient parser (the htmlparser2 kind) that tries to interpret the document even with badly closed tags, and returns counts, not a list of spec-conformance errors.
Yes. Those are the 14 void elements in the HTML spec (area, base, br, col, embed, hr, img, input, link, meta, param, source, track, wbr) and each occurrence adds 1 to the tag total, even with no matching </tag> in the text.
Yes, it is one of the two accessibility checks: an image lands on the list when the alt attribute is missing or ends up empty after trimming whitespace, which covers both an oversight and an alt="" left blank on purpose.
No structure has been detected yet.
No structure has been detected yet.
No headings were found in the analyzed HTML.
No images without alt text were detected.
No links were detected.
Structural analysis runs locally. The HTML is not sent to the server and is not rendered to execute scripts.