Schema Validator

Stack

• JSON-LD parser + microdata fallback parser
• Schema.org core vocabulary validator (kept in sync with the public Schema.org release)
• Google rich-results rule engine (FAQ, HowTo, Product, Review, Recipe, Article, Breadcrumb, Event, Job)
• Cross-reference checker for `@id` graph linking

Training / evaluation data

Not a learned model — rules track the Schema.org spec and Google's developer docs directly. Refreshed when either source publishes updates.

Known limitations

• Brand-new schema types added since our last spec sync may show as 'unknown' until we update
• Google's rich-result eligibility depends on factors beyond markup (E-E-A-T, page quality, freshness) that the validator can't assess
• Some validators flag unused-but-allowed properties as warnings; we surface those as informational, not errors
• We don't render JS to extract schema injected at runtime — submit the rendered HTML for SPAs