SOURCE CHECKING, ON YOUR DEVICE
HTML Validator
Find source mistakes before publishing. Check pasted HTML or a local file, jump to a problem, and take an actionable report back to your editor.
A practical source checker. Syntax errors, review hints, and coverage limits are reported separately. This is a limited local check, not full HTML conformance validation or a W3C certificate.
Your HTML source
Up to 250,000 UTF-16 units. Line endings become LF. Source is checked as text and is never rendered or executed. Ctrl/Cmd + Enter runs the check.
0 / 250,000 units
.html, .htm, or .txt, up to 1 MB (1,000,000 bytes). Opening a file replaces the editor. No upload.
Source diagnostics
Ready. Paste HTML or load the example to begin.
0 Syntax errors
0 Review hints
0 Coverage notes
Keep the report
Exports include all listed diagnostics and short source excerpts, regardless of the filter. Review excerpts before sharing.
View or manually copy the report
How to check and fix HTML source
- Paste the HTML delivered to a browser or open a UTF-8 file. Select HTML fragment for a CMS block or snippet that has no document doctype.
- Choose Check HTML. Start with syntax errors, then review the contextual hints. Select a line-and-column button to focus and select that source range.
- Edit the source and check again. Changes clear the previous report so its locations cannot point into a newer version.
What the results mean
Syntax error: a directly observed source problem, such as duplicate attributes, malformed attribute syntax, or an end tag for a void element. Browsers often recover from these problems; a page looking correct does not prove its source is correct.
Review hint: a source pattern needing context, such as unbalanced nesting, duplicate literal IDs, missing image alternatives, or document metadata. These are heuristics. The checker does not construct a browser DOM, calculate accessible names, or verify every HTML content model.
Coverage note: a region or feature that was not fully checked. Scanning stops at SVG, MathML, templates, noscript, certain legacy contexts, processing instructions, template markers, or complex script escapes. A partial report clearly identifies where it stopped. Check ordinary HTML regions separately and use a fuller checker for the complete document.
Optional tags are part of HTML
HTML permits some tags to be omitted under specific conditions. For example, <ul><li>One<li>Two</ul> can omit both list-item end tags. This tool recognizes common implied closures and does not treat every omitted tag as an error. It does not verify all omission conditions, table insertion modes, or browser recovery rules. Legal uppercase names, unquoted attribute values, and <br /> are accepted.
Scope and privacy
This tool makes no requests with your source and does not save it in browser storage. Files stay on your device. Submitted HTML cannot load images, run scripts, or create frames here. The site's shared page assets and analytics are independent of the checker; this tool does not send them source or reports.
HTML source is limited to 250,000 UTF-16 units, 20,000 tokens, 200 levels of source nesting, and 200 listed diagnostics. Columns count Unicode code points, with a tab counting as one. Entity-containing IDs are excluded from ID comparison. CSS, JavaScript, character-reference validity, URL availability, XML/XHTML, framework templates, full accessibility, and rendered behavior are outside the checks.
For broader checks, use the Nu Html Checker and review the HTML syntax and HTML parsing rules. These links open external sites. Your source is never attached to them.