openalex - SKILL.md Agent Skill

name: openalex user_invocable: true description: Query and analyze scholarly literature using the OpenAlex API (240M+ works). Use for literature searches, finding papers by author/institution, citation analysis, and bibliometric queries.

openalex — Scholarly Literature Search

Query the OpenAlex API for academic papers, authors, institutions, and citation data. No API key required.

When to Use

"find papers by", "search for papers on", "who cites this"
Literature search, citation analysis, bibliometric queries
Finding works by author, institution, or topic

Client

from scripts.openalex_client import OpenAlexClient

client = OpenAlexClient(email="user@example.edu")  # email → 10x rate limit

Script paths are relative to this skill's directory.

The client handles rate limiting, retries with exponential backoff, and cursor-based pagination.

Key Patterns

Search works

results = client.search_works(search="CRISPR", filter_params={"publication_year": ">2020", "is_oa": "true"}, sort="cited_by_count:desc")

Two-step entity lookup (always get ID first, never filter by name)

author = client.get("/authors", params={"search": "Jennifer Doudna", "per-page": 1})
author_id = author["results"][0]["id"].split("/")[-1]
works = client.search_works(filter_params={"authorships.author.id": author_id})

Get entity by external ID

work = client.get_entity("works", "https://doi.org/10.1038/s41586-021-03819-2")
author = client.get_entity("authors", "https://orcid.org/0000-0003-1613-5981")

Batch DOI lookup (up to 50 per request)

works = client.batch_lookup("works", doi_list, "doi")

Paginate large results (cursor-based)

all_papers = client.paginate_all("/works", params={"search": "quantum computing", "filter": "publication_year:2022-2024"}, max_results=5000)

Aggregate by field

by_year = client.group_by("works", "publication_year", filter_params={"authorships.author.id": author_id})

Filter Syntax

Single:     publication_year:2023
AND:        publication_year:>2020,is_oa:true
OR:         type:journal-article|book
NOT:        type:!paratext
Range:      publication_year:2020-2024
Greater:    cited_by_count:>100
Both inst:  authorships.institutions.id:ID1+ID2    (AND)
Either:     authorships.institutions.id:ID1|ID2    (OR)

Common filters: publication_year, is_oa, cited_by_count, type, authorships.author.id, authorships.institutions.id, primary_location.source.id, topics.id, has_doi.

Attribution

OpenAlex client adapted from K-Dense-AI/claude-scientific-skills (MIT License).