I'm working on a project where we want to make content harvestable by Zotero so researchers can easily add items to their library. Needs to support multiple items on a page so that, e.g. from a search results page with 50-100 items a researcher can import all or select some of the items to add to their Zotero library.

What's the current recommended solution for this? On the documentation page for exposing metadata https://www.zotero.org/support/dev/exposing_metadata the only options that can work for multiple items on a page are COinS and unAPI, but both of their websites are dead, which I find troubling. It sounds like it would still work for Zotero - but it's probably not going to be useful for anything else. I read the conversation about unAPI ( https://forums.zotero.org/discussion/50327/is-unapi-dead-because-their-website-sure-seems-to-be ) and wondered if there were any updates on this. Are there any methods for exposing metadata that involve more current technologies, like JSON-LD ?

For what it's worth, I can get MARCXML for most of the content I'm working with; if my only choices are COinS and unAPI, then I would pick unAPI. I just want to make sure there's not a better solution before I start implementing.

Thanks for any advice!