Many thanks to everyone who suggested OpenAlex! I've added it to the MCP.
I also added PubMed, a better README, some optimization files for Smithery,
and some files that should allow you to easily contribute to the project if
you use Claude Code (as I started using this morning to get through the
Smithery optimization... my first time paying for an LLM).
It's running a bit heavy while checking three different APIs. Recall is
noticeably better and noticeably slower. It seems like something I did also
made me have to more explicitly tell Claude to use it now, e.g. "find
articles about" instead of "tell me about." I'm working on some performance
tuning.
Thanks again,
*Thomas Scharff*
<https://library.wheatoncollege.edu/prf.php?account_id=205579> (he/him/his)
*Director of Discovery & Access*
Madeleine Clark Wallace Library <https://library.wheatoncollege.edu>, G22
Wheaton College <https://wheatoncollege.edu>
26 E. Main Street, Norton, MA 02766 <https://goo.gl/maps/d5JvoKZUEXy>
(508) 286-5601 <+15082865601>
On Fri, Oct 31, 2025 at 4:00 PM Park, Sarah <gpark1_at_illinois.edu> wrote:
> [like] Park, Sarah reacted to your message:
> ________________________________
> From: Code for Libraries <CODE4LIB_at_LISTS.CLIR.ORG> on behalf of Eric
> Lease Morgan <00000107b9c961ae-dmarc-request_at_LISTS.CLIR.ORG>
> Sent: Friday, October 31, 2025 7:25:27 PM
> To: CODE4LIB_at_LISTS.CLIR.ORG <CODE4LIB_at_LISTS.CLIR.ORG>
> Subject: Re: [CODE4LIB] DOI Citation Verifier MCP Server [open alex]
>
> I have had pretty good success with Open Alex; given a list of DOIs, I
> have used Open Alex to augment citations with robust bibliographic values,
> download the associated PDF files, and measure the impact of the cited
> documents.
>
> For example, suppose I have a CSV file and one of the columns is named
> "doi" (i.e. survey-results.csv), then I can feed that CSV file to
> dois2metadata.py and it will output a second CSV file, metadata.csv. This
> second CSV file includes robust values for authors, titles, dates, and more
> importantly, locations where the full text can be gotten.
>
> Once I get this far, I loop through the metadata file to get Open Alex
> "works" files. These are JSON files with even more detail including author
> titles, author affiliations, lists of associated controlled vocabulary
> terms, and sometimes abstracts. I use dois2works.py for this purpose, and
> it creates a directory of "works" files. I can use another script,
> cache-pdfs.py, which loops through the same metadata file and caches PDF
> files, if they are available.
>
> Open Alex also has "source" files, and these files include citation
> metrics; number of times cited, h-index, etc. I can get these source files
> using works2sources.py, and the results are saved in the sources directory.
>
> Finally, I can combine the "works" files with the "source" files with to
> create yet another CSV file, bibliometrics.csv, which includes rudimentary
> identifiers and various citations scores.
>
> How might I use this suite of software? Well, I might use the Web Of
> Science API to search... Web Of Science. This results in a CSV file of
> authors, titles, dates, and sometimes DOIs. I can then feed the CSV file to
> this suite of software to augment the results and actually get the
> articles. Once I get abstracts and/or articles, I can then create a data
> set against the whole for the purposes of analysis (distant reading). Thus,
> using this system, I can literally read hundreds if not thousands of
> articles.
>
> In summary, given a list of DOIs, it is possible to use Open Alex to get
> robust bibliographic data, the associated full text, as well a
> bibliometrics. Moreover, there are no dollar costs associated with Open
> Alex. It is really "free as in free beer".
>
> All of the scripts and intermediate files created by this software suite
> are temporarily linked below. And yes, I know people do not like obfuscated
> links, but if I didn't use a shortener, then the links would be
> embarrassingly long:
>
> * survey-results.csv -
> https://urldefense.com/v3/__https://bit.ly/3JpI6Xk__;!!DZ3fjg!49eaE4LOUsqOPMA3PzEFSym1-M_fguLAy-Wepijd4ofND7ZMvAtZVSr1-2hCTNr4mrjWKmyRVKOFIHITegufIMwPZGhSosuclQ$
> * dois2metadata.py -
> https://urldefense.com/v3/__https://bit.ly/3Wu1TI5__;!!DZ3fjg!49eaE4LOUsqOPMA3PzEFSym1-M_fguLAy-Wepijd4ofND7ZMvAtZVSr1-2hCTNr4mrjWKmyRVKOFIHITegufIMwPZGjBaz1bLA$
> * metadata.csv -
> https://urldefense.com/v3/__https://bit.ly/3JBI1Qa__;!!DZ3fjg!49eaE4LOUsqOPMA3PzEFSym1-M_fguLAy-Wepijd4ofND7ZMvAtZVSr1-2hCTNr4mrjWKmyRVKOFIHITegufIMwPZGhr4henMA$
> * dois2works.py -
> https://urldefense.com/v3/__https://bit.ly/43agPPb__;!!DZ3fjg!49eaE4LOUsqOPMA3PzEFSym1-M_fguLAy-Wepijd4ofND7ZMvAtZVSr1-2hCTNr4mrjWKmyRVKOFIHITegufIMwPZGjlEfpqfA$
> * directory of works files -
> https://urldefense.com/v3/__https://bit.ly/3JnJDg__;!!DZ3fjg!49eaE4LOUsqOPMA3PzEFSym1-M_fguLAy-Wepijd4ofND7ZMvAtZVSr1-2hCTNr4mrjWKmyRVKOFIHITegufIMwPZGhgfvSxCA$
> * cache-pdfs.py -
> https://urldefense.com/v3/__https://bit.ly/4oMgMBl__;!!DZ3fjg!49eaE4LOUsqOPMA3PzEFSym1-M_fguLAy-Wepijd4ofND7ZMvAtZVSr1-2hCTNr4mrjWKmyRVKOFIHITegufIMwPZGjRnEfCJQ$
> * works2sources.py -
> https://urldefense.com/v3/__https://bit.ly/432xRPm__;!!DZ3fjg!49eaE4LOUsqOPMA3PzEFSym1-M_fguLAy-Wepijd4ofND7ZMvAtZVSr1-2hCTNr4mrjWKmyRVKOFIHITegufIMwPZGjhI0d4ug$
> * sources -
> https://urldefense.com/v3/__https://bit.ly/3JqOVrC__;!!DZ3fjg!49eaE4LOUsqOPMA3PzEFSym1-M_fguLAy-Wepijd4ofND7ZMvAtZVSr1-2hCTNr4mrjWKmyRVKOFIHITegufIMwPZGinJxuwtA$
> * works2sources.py -
> https://urldefense.com/v3/__https://bit.ly/4oNO5UW__;!!DZ3fjg!49eaE4LOUsqOPMA3PzEFSym1-M_fguLAy-Wepijd4ofND7ZMvAtZVSr1-2hCTNr4mrjWKmyRVKOFIHITegufIMwPZGiKVbLxYQ$
> * works2matrix.py -
> https://urldefense.com/v3/__https://bit.ly/439hmku__;!!DZ3fjg!49eaE4LOUsqOPMA3PzEFSym1-M_fguLAy-Wepijd4ofND7ZMvAtZVSr1-2hCTNr4mrjWKmyRVKOFIHITegufIMwPZGi1AVU38Q$
> * bibliometrics.csv -
> https://urldefense.com/v3/__https://bit.ly/43KgXFe__;!!DZ3fjg!49eaE4LOUsqOPMA3PzEFSym1-M_fguLAy-Wepijd4ofND7ZMvAtZVSr1-2hCTNr4mrjWKmyRVKOFIHITegufIMwPZGhjkhgHcg$
>
> (On the other hand, your email provider may very well obfuscate the links
> a different way, whether you like it or not. Dumb, if you ask me, but that
> is a different discussion.)
>
> --
> Eric Lease Morgan
> Librarian Emeritus, University of Notre Dame
>
Received on Wed Nov 05 2025 - 09:29:35 EST