CDIF metadata recommendations are packaged in a set of profiles. Each profile includes:
A Content model that specifies the information expected to be included in any metadata record, with required, recommended and optional content items. Each profile description starts with a description of its content model, e.g. for the core profile
A JSON-LD serialization for that content using the Schema.org vocabulary to define the fields in a metadata record, e.g. Core profile JSON-LD serialization, and an implementation using the DCAT rdf vocabulary
Tools for validating metadata instance documents, including JSON schema, SHACL rules, a JSON-LD framing document, and python code to run validation.
Each CDIF profile is housed in a github repository containing an implementation guidance document, JSON schema, SHACL rules, a JSON-LD framing document, and a subdirectory containing examples.
An additional critical part of the recommendations is description of Workflows to publish CDIF metadata so that is can be found and indexed by search providers using standard web technology.