CORE Open Access Paper Search
Pricing
from $2.00 / 1,000 paper fetcheds
Search and extract open access academic papers from CORE -- the world's largest aggregator of open access research with over 300 million metadata records and 40+ million full-text papers. Filter by keyword, year range, and language.
Your free CORE API key. Register at https://core.ac.uk/services/api to get one.
Keywords to search across titles, abstracts, and full text. Supports Boolean operators (AND, OR, NOT) directly.
One-pick research intent that sets sensible defaults for the gate, dedupe, filters, and monitoring. Your own explicit settings always override the preset. literature-review: read-ready, de-duplicated, grade B+ with PDF verification. state-of-the-art: recent (<=3y) and already cited (>=25). historical-foundations: highly cited (>=100). thesis-sources: theses only. monitoring: emit only changed papers across runs.
Phrases that must appear verbatim (each is matched as an exact phrase).
Restrict the main query to paper titles instead of full text.
Only return papers with at least this many citations.
Filter by language code (e.g., 'en' for English, 'de' for German).
Only return papers that have a downloadable full text PDF.
Keep only papers whose document type matches one of these (e.g. 'research', 'thesis', 'preprint'). Applied after fetch.
Keep only papers hosted by a repository whose name contains this text (e.g. 'arXiv', 'Strathclyde'). Applied after fetch.
Maximum number of papers to return (up to 500).
none: no check. url-present: flag whether a download URL exists (no network call). head-check: live HEAD request per PDF to confirm it is downloadable now (adds one request per paper).
Acceptance criteria object, e.g. { "minGrade": "B", "requirePdf": true, "minCitations": 5, "maxAgeYears": 8, "excludeDuplicates": true }.
When the quality gate is set, exclude dropped papers from the dataset (still counted in the corpus summary).
none: off. doi: DOI exact match. strict: DOI or normalized title+author+year. fuzzy: strict plus near-identical title detection.
Name a watchlist to track papers across runs. Each paper is flagged NEW the first time it appears, SEEN afterwards, with changedFields (citationCount, retrievability, becameReadable). Leave blank for one-off searches.
flag-all: emit every paper with its watchlist status. new-only: emit only papers new since the last run. changed-only: emit only new or changed papers.
Optional. A second query to benchmark against the main query. Emits a 'benchmark' record comparing the two corpora on health score (PDF availability, readiness, de-duplication). Runs a second CORE search.
papers: one record per paper plus a corpus-summary digest. corpus: suppress per-paper rows and return only the run-level corpus-summary digest — charged as a single event. benchmark: compare your query against compareQuery on corpus health and return only a head-to-head comparison table — charged as a single event (requires compareQuery).
