The thing that confuses me is: > A keyword is "known" if it is defined by a vocabulary listed in the schema's meta-schema (identified by the value in $schema).

The thing that confuses me is: > A keyword is "known" if it is defined by a vocabulary listed in the schema's meta-schema (identified by the value in $schema). { "$schema": "https://json-schema.org/draft/2020-12/schema", So, if a schema identifies as conforming to the 2020-12 meta-schema, why would a new "database-field-id" keyword in the 2025 upgrade be an issue? The 2025 parsers should know that it's not a keyword in the 2020-12 meta-schema, and treat it the same as they always did, shouldn't they? But also - as an alternate to underscore prefixes, could new keywords just reuse the "$" prefix as in "$schema"?