Open data

Everything on this site is available as structured data under the CC BY 4.0 licence. Please cite the page URL and the version.

• corpus.json: every summary with its claims, interpretations, limits and full source records.
• claims.jsonl: one sourced claim per line, suited to retrieval systems.
• sources.json: the source registry.
• feed.xml: Atom feed of new and updated summaries.
• llms.txt: a plain-text map of the site for language-model tools.

Publishing data openly makes it easier to find and reuse. It does not guarantee that any search engine or AI system will use it.