...
Metadata is provided in the MODS schema (http://www.loc.gov/standards/mods/) as extended in the Harvard LibraryCloud Schema (URL)
Base URI
The the service's base address is: https://api.lib.harvard.edu/oai
ListSets
The ListSets verb will return a list of harvestable OAI sets defined as LibraryCloud collections. At this time, the LibraryCloud ListSets verb does not support a resumptionToken, so the response will always include the first 10 collections only.
https://api.lib.harvard.edu/oai/?verb=ListSets
ListRecords
The ListRecords verb will return full metadata records for a given LibraryCloud collection. Results can be returned as MODS or DC.
https://api.lib.harvard.edu/oai/?verb=ListRecords&metadataPrefix=mods&set=scores
https://api.lib.harvard.edu/oai/?verb=ListRecords&metadataPrefix=oai_dc&set=scores
ListIdentifiers
The ListIdentifiers verb will return the identifiers of records in a given LibraryCloud collection.
https://api.lib.harvard.edu/oai/?verb=ListIdentifiers&metadataPrefix=mods&set=scores
Identify
The Identify verb will return information about Harvard's LibraryCloud repository.
https://api.lib.harvard.edu/oai/?verb=Identify
GetRecord
The GetRecord verb will return an individual metadata record from the LibraryCloud repository. Required arguments specify the identifier of the item to be requested and the format of the metadata that should be included in the record.
https://api.lib.harvard.edu/oai/?verb=GetRecord&metadataPrefix=mods&identifier=000796856 (using Aleph ID)
https://api.lib.harvard.edu/oai/?verb=GetRecord&metadataPrefix=oai_dc&identifier=990007968560203941 (using Alma ID)
ListMetadataFormats
The ListMetadataFormats verb will return the metadata formats available from the LibraryCloud repository. An optional argument restricts the request to the formats available for a specific item.
...
https://api.lib.harvard.edu/oai/?verb=ListMetadataFormats&identifier=990007968560203941 (why doesn't this work?)
Resumption tokens
The OAI requests for ListSets, ListRecords, and ListIdentifiers retrieve 10 results at a time, so the repeated use of a resumptionToken in the request URL is needed to retrieve the full amount of data.
...