Is your schema actually valid, not just present?

Google's tool checks eligibility for specific search result features. This checks the fields that matter for AI answer engines specifically: whether an Organization has a name, whether an Article has an author and date, whether a FAQPage actually has usable questions. It also tells you which AI-relevant types (Organization, Article, FAQPage) are missing from the page entirely, not just whether what you have is valid.

An error is a required field the schema type cannot function without, for example an Organization with no name, or a FAQPage with no mainEntity. A warning is a recommended field: present or not, the schema still validates, but adding it (a logo, a datePublished, an author) strengthens the trust and specificity signals AI engines look for.

It is flagged as a warning, not a hard error. Some setups use a shorthand or a URL variant of the schema.org context. If your schema otherwise renders correctly and search tools accept it, a slightly different @context string is usually cosmetic, but the checker flags it so you can confirm it resolves the way you expect.

It fetches the live HTML of the page you enter and pulls out every <script type="application/ld+json"> block on it, then runs each one through the same validation as paste mode. It does not execute JavaScript, so schema injected client-side after page load will not show up.

It shows whether the whole page (across every block) declares Organization, Article/BlogPosting, or FAQPage schema, the three types most consistently rewarded by AI answer engines. A page can have perfectly valid schema and still be missing one of these entirely, which the per-block results alone would not surface.

No. Valid, complete schema makes a page easier for a crawler to parse correctly, which is necessary but not sufficient. Whether an AI assistant actually surfaces and cites your page also depends on off-site signals like where else your brand is mentioned, which this tool cannot see.

Pasted JSON-LD is validated entirely in your browser and never leaves it. URL mode fetches the target page server-side (browsers cannot fetch other origins), runs the check, and returns the result without storing it.