literature-search-openalex

literature-search-openalex lets an agent query the OpenAlex scholarly database for research papers, authors, institutions, topics, sources, publishers, funders, geo-locations, and keywords. It solves the problem of programmatic academic literature discovery and bibliometrics — searching papers, resolving DOIs, finding an author's publications, aggregating citation counts and h-index/impact metrics, exploring the research taxonomy, and downloading open-access PDFs — through a single bundled CLI (scripts/openalex_cli.py) run via uv, which handles retries and rate limiting.

The CLI exposes resolve (name to ID candidates), get (full metadata for one entity), filter (search/filter with --filter, --search, --sort, --select, --group-by, --sample), download-pdf, and rate-limit commands across many entity types. The skill enforces disciplined behavior: always resolve a name to an ID before filtering, never fabricate OpenAlex IDs or DOIs, use the CLI rather than raw curl/urllib, keep output small with --select and small page sizes, and list the source URLs of all papers used. It documents rate limits and per-operation costs for the free 'polite pool' versus the Premium API. Notably, it requires the operator to notify the user to check OpenAlex terms and the license of retrieved papers, records that notification in a .licenses file, and uses a separate credentials skill's safe protocol to check for and request the optional OPENALEX_API_KEY (stored in a home-directory .env) rather than embedding secrets.

It targets researchers, data scientists, and science-focused agents performing systematic literature review, citation analysis, author/institution profiling, or DOI resolution. Extensive reference files document the queryable fields for works, authors, institutions, sources, topics, publishers, funders, geo/language, and type enumerations.