Scholarly literature databases for AI-assisted literature search

A literature review usually means running the same search in five places and reconciling the results by hand. Kahubi connects to the open scholarly infrastructure: OpenAlex and Crossref for metadata and citations, Europe PMC and bioRxiv for life-science papers and preprints, arXiv for physics, computing and mathematics, DOAJ for open-access journals, HAL, OpenAIRE and Norway’s Cristin for European research output, ERIC for education research, ORCID for researcher identities and zbMATH for mathematics.

The assistant writes the query for each database, merges duplicates and cites every paper by DOI or repository identifier. Preprints are marked as not peer reviewed. Papers you want to keep go straight into your Kahubi library.

12 sources you can connect

Questions researchers ask these sources

  • “Which institutions published the most on microplastics in freshwater between 2019 and 2024?”

    Institution counts with OpenAlex entity IDs, so the ranking can be reproduced.

    OpenAlex
  • “Resolve these ten DOIs and give me complete metadata for my reference list.”

    Titles, authors, journals, years and publishers straight from the DOI registry.

    Crossref
  • “Look up the public ORCID record for this researcher and list their works since 2020.”

    Works, affiliations and funding from the public record, respecting the researcher’s visibility settings.

    ORCID
  • “What are the most recent arXiv preprints on retrieval-augmented generation for scientific question answering?”

    Preprints with title, authors, submission date, category and abstract, each flagged as not yet peer reviewed.

    arXiv
  • “Find open-access papers since 2020 on microbiome and depression, sorted by citations.”

    Records with PMID, DOI, journal, year and citation count, then the full text of any open-access article.

    Europe PMC
  • “Find open-access datasets from Swedish institutions on urban heat islands published since 2018.”

    Research products with type, year, access rights, persistent identifier and hosting repository.

    OpenAIRE Graph

Frequently asked questions

Does this replace Scopus or Web of Science?
For finding papers it covers most of the same ground: OpenAlex indexes over 250 million works. For formal bibliometric studies that require a specific licensed index, keep using it; Kahubi is the faster way to search, read and cite.
Can it read the full text of papers?
Where the full text is open (Europe PMC, arXiv, bioRxiv, HAL and open-access journals) the assistant can read it and quote passages. For paywalled papers it works from the abstract and metadata unless you add the PDF to your library.
How do I avoid made-up references?
Every reference in an answer comes from a record the assistant retrieved, with its DOI or identifier. You can open each one, and Kahubi’s source check flags claims that the cited paper does not support.

How MCP grounds AI answers in real records

Search scholarly literature sources in one research chat

Free to start. Connect a source in one click and get answers that cite the records they came from.