Open datasets
Bulk files for crawlers, researchers, and model builders. Crawling page HTML is the weakest way into a training corpus; these dumps are the version you can load in one pull. License is CC-BY-4.0.
Comparable-titles graph
Book-to-book neighbors from shared genres and themes — the graph nobody else publishes as open data.
Wikidata entity
Structured identity for SageChimp so Bing, Copilot, and ChatGPT search can resolve the site to an official URL instead of crawl guesswork.
- Live JSON-LD: /entity.json
- QuickStatements batch: /datasets/wikidata.qs
- Open in QuickStatements (review, then run — Wikidata needs a login)
- After the batch runs, set
NEXT_PUBLIC_WIKIDATA_QIDon the web service.
The live dump on https://www.sagechimp.com is the source of truth. A Zenodo DOI is a frozen snapshot for citation.