Thanks for sharing, Ricardo! And - I realize you never got any kind of reply on your thread - sorry about that!
We at Artefactual have not tested any OAI harvesters, but it is definitely possible. Several of our UK AtoM users are working with members of the ArchivesHub to use OAI-PMH to pull in AtoM EAD 2002 XML descriptions. You could try reaching out to one of their members for more information on the harvester they use.
Finally, with a quick search I also found some information on 2 open source harvester tools mentioned in the cultural heritage community - one from the Digital Public Library of America (DPLA), and one from OCLC:
As I have mentioned in other threads, AtoM also has basic support for REST-like API endpoints:
Additionally, Elasticsearch (the search index library we use in AtoM) has its own API, and we also use Elastica, a PHP client for Elasticsearch, which also has an API. In the current 2.4.1 version of AtoM we are still using ES 1.7, but in the 2.5 release, we will be upgrading AtoM to use Elasticsearch 5.6 (or later if we have time to review it again against the latest ES release). See:
I have mentioned in previous threads that the University of British Columbia Library has built a portal site (
Open Collections) that pulls in content from multiple different applications - for AtoM, they were pulling this data directly from the ES index. I think they've made some changes (the same collection is no longer included), but we previously included this integration in slides 13-18 of the following deck:
Additionally, another user in the forum has shared a front end search interface they have made that searches across 2 different AtoM sites using the ES indexes - see:
You would need to develop your own way of consuming this data. However, there are many resources available online that will introduce APIs and how to work with them that can point you in the right direction.
Good luck!