Semantic HTML checker

Clean markup is what makes a page parseable: one H1, a logical outline, landmarks, real link text. This checker covers the layer everything else builds on.

Check your siteHow scoring works

5 checks · 6% of the Slurp Score

Why it matters for AI search

The H1 declares your topic. Headings define where retrieval systems split your page into chunks. Landmarks separate content from navigation. The lang attribute picks your language index. All basic, which is why failing here is so unnecessary.

It counts double now. Crawlers parse this structure into indexes, and browsing agents like ChatGPT-User operate live pages for users. An agent that cannot tell your content from your navigation serves your visitors badly.

What GetSlurp checks

This list comes straight from the audit engine, so it always matches what the extension and MCP server actually run.

semantic.single-h1Graded

The H1 is the strongest single statement of what a page is about; zero or multiple H1s blur the topic for extraction.

Fix: Use exactly one H1 that states the page topic.

semantic.heading-orderGraded

Skipped heading levels break the outline that retrieval systems use to segment the page.

Fix: Do not skip heading levels (H2 to H4); keep the outline sequential.

semantic.landmarksGraded

Landmark elements (main, nav) tell parsers and browsing agents which part of the page is content versus chrome.

Fix: Wrap the primary content in <main> and navigation in <nav>.

semantic.langGraded

The lang attribute tells engines which language index and answers the page belongs in.

Fix: Set lang on the html element (e.g. <html lang="en">).

semantic.link-textGraded

Descriptive anchors tell machines what the target is; "click here" carries zero information.

Fix: Rewrite generic anchors ("click here", "read more") to describe their destination.

Try GetSlurp free

Audit any page with 60+ checks across 12 categories. No sign-up, runs on your machine.

Get the extension

Related checkers

All 12 categories