Use data transformation routines extracted from LUX (https://github.com/project-lux/data-pipeline) to retrieve records from a variety of data sources, then transform those records into Linked Art, to then provide access in code, over HTTP, or via MCP for LLM consumption. It adds a searcher class that allows (simple) searching of the data sources.
It was developed during a workshop organized to explore the potential of linked data and large language models for cultural heritage.
- National Library of Spain
- National Library of France
- National Library of Japan
- German National Library
- OCLC's FAST
- Global Biodiversity Information Facility (GBIF)
- GeoNames
- Getty AAT, ULAN and TGN vocabularies
- Homosaurus
- Library of Congress Subject Headings
- Library of Congress Name Authority File
- Yale's LUX
- Nomisma
- ORCID
- Pleiades
- ROR
- SNAC
- VIAF
- Wikidata
- Who's on First