← clify home · all task guides · agent guide

Science and research data APIs for AI agents

Search papers, datasets, trials, genes, proteins, chemistry and research records through public JSON endpoints.

Use these verified site commands for reproducible literature and scientific-data lookups. Follow each source page for required identifiers, filters and example requests.

Verified sources

arxiv

arxiv.org

arXiv.org, the open-access preprint archive. This CLI browses category listings (new, recent, monthly), daily catch-up, archive overviews, yearly stats, the category taxonomy and RSS feeds, and fetches paper abstract pages and BibTeX citations.

Machine-readable commands · 10 verified commands

pubmed-ncbi

pubmed.ncbi.nlm.nih.gov

PubMed (NCBI biomedical literature database): search articles by keyword, author or field-tagged query, with filters, sorting and pagination, and fetch full abstract records, MEDLINE text exports and PMID lists. All commands read the server-rendered search page.

Machine-readable commands · 4 verified commands

clinicaltrials

clinicaltrials.gov

ClinicalTrials.gov (NIH/NLM registry of over 600k clinical studies). The CLI searches studies, fetches full study records and their version history, autocompletes conditions, interventions, sponsors and facilities, and returns registry statistics and field metadata, all through the site's internal /api/int endpoints.

Machine-readable commands · 8 verified commands

semanticscholar

semanticscholar.org

Semantic Scholar, an AI-powered search engine for scientific literature. The CLI searches and looks up papers, authors, citations, references and recommendations through the official Semantic Scholar Academic Graph and Recommendations APIs, and also offers the web app's title autocomplete.

Machine-readable commands · 4 verified commands

uniprot

uniprot.org

UniProt, the Universal Protein Resource. This CLI wraps the public rest.uniprot.org JSON API to search and fetch UniProtKB protein entries, sequences, publications, version history, taxonomy, keywords, diseases, UniRef clusters and proteomes, plus AlphaFold structure predictions.

Machine-readable commands · 18 verified commands

ebi

ebi.ac.uk

EBI Search (EMBL-EBI): keyword search across all EMBL-EBI data resources such as UniProt, PDBe, Ensembl, Europe PMC, ChEBI and Reactome, with entry lookup, cross-references, gene/protein identification and FASTA sequence download.

Machine-readable commands · 11 verified commands

pubchem-ncbi

pubchem.ncbi.nlm.nih.gov

PubChem (NCBI) open chemistry database. This CLI searches compounds, genes, bioassays and literature, runs structure searches, fetches compound, gene, assay and element records, exports 2D/3D structures, and browses the periodic table, classification trees and data sources.

Machine-readable commands · 19 verified commands

zenodo

zenodo.org

Zenodo (CERN's open research repository). This CLI searches communities and the records inside them, and fetches record metadata, versions, formatted citations, citing works, community members, subcommunities and collection records.

Machine-readable commands · 12 verified commands

inspirehep

inspirehep.net

INSPIRE-HEP is the high-energy physics literature database. This CLI searches and fetches papers, citations, references, BibTeX/LaTeX exports, authors, jobs, conferences, seminars, institutions and experiments through INSPIRE's public JSON API.

Machine-readable commands · 20 verified commands

rcsb

rcsb.org

RCSB Protein Data Bank (rcsb.org): search 3D macromolecular structures by keyword, attribute or 3D shape, and fetch entry chains, sequence annotations, UniProt alignments, version history, ligand definitions and coordinate/FASTA files.

Machine-readable commands · 17 verified commands

How to use this guide

Open a source page for its command list, or use its .json link as a machine-readable schema. Start with the command's example URL, then narrow results with the documented parameters.