Zotero 101 - Obsidian Community talks

I have roughly sketched out a table of contents of what I could present in the Zotero 101 talk, but I’d like to use this thread to collect requests of what you as an audience would like me to cover or any questions you already have. For now, I’ve assumed that most of the people interested in this talk are not necessarily academics, but are at least interested in the PDF annotation workflow or want to know what a reference manager is useful for. If you want me to add something else, have a specific use case in mind or if you want me to demonstrate something, please let me know in the comments just in case I need to prepare it in advance.


  • Why do you need a reference manager?
  • Zotero GUI and what you can do out of the box with it
    • Why do you need plugins and which ones do you need?
  • Basic setup
    I’ll start with a fresh profile of Zotero and walk you through the whole process. Depending on the available time, I might choose a few points below where you can follow along.
    • Zotero
      • Where does it store its data?
      • Ways to add new items
      • Collections and tags
      • Installing plugins
    • Setting up Zotfile
      • Managing attachments (renaming rules)
      • Extracting annotations
    • Setting up Better BibTeX
      • Citekey management
      • Exporting bibliography for use with other tools
    • Setting up Mdnotes
      • File configuration (split vs multiple files)
      • Templates
      • Field customization
  • Q&A


  • Zotero v5.0.96 (Installed)

And the xpi of the following plugins downloaded to your computer:

  • Zotfile v5.0.16 (Download here)
  • Better BibTeX v5.2.144+(Download here)
  • Mdnotes v0.2.0 (hopefully I’ll manage to release it before the talk, download here)

Note: If you can only make it from 16:30 onwards, try to have the holy trinity of plugins already installed if you want to follow along.

Date and time

Please vote for a slot that works for you using when2meet: https://www.when2meet.com/?11609682-Sy5UA

Based on the votes and schedule, the talk will take place on 2021-04-24T14:00:00Z (CET, click on the date for your timezone). I’ll be aiming for an hour, but it might be that we need a bit longer for discussions or questions at the end.

Presentation and Recording




I’m interested in getting information from journal articles in Zotero into Obsidian and from there into Scrivener and/or LibreOffice or MS Word

1 Like

I’d personally prefer some extra focus on mdnotes, as it’s the only piece I haven’t put into the puzzle yet :yum:

I’m only a hobbyist in this realm, and very new to Zotero. In fact I started using it after discovering Obsidian. My problem - The majority of Zotero tutorials focus heavily on very specific uses - mostly academic.

Although I’ve really only been saving web content (almost purely as .pdf’s), I do plan on building a Goodreads-style media shelf to track things I’ve consumed, or want to consume someday. This would include ALL media - books (fiction and nonfiction), music, film, television, et cetera.
Zotero is pretty straightforward for research. It would be neat to see if others have thoughts on how those more …casual… things could be organized, annotated, and eventually extracted once finished.

That, and since I’ve abandoned folders for most purposes, Zotero Collections scare me just a little bit :slight_smile:

I have managed to cobble together Zotfile and Mdnotes into what I feel is a pretty usable output, so I’m not quite as interested in that. However, it took me no end of fiddling and trial-and-error to get there, so I certainly understand this talk could be Mdnotes heavy, and with good reason. I suspect that will be on the minds of most that participate. :wink:

I guess the bottom line is I (personally) would look forward to the Zotero/Obsidian specific content as much or more than the extraction tools. Basically, it would just be cool to see how other real people have used Zotero.

1 Like

The thing I’m most stuck on is changing the output format from the default. From the documentation it looks really technical and I’m a little scared of it.

Thank you for doing this!
Will this be recorded?

@Kullenej I’m not sure yet! I’ll announce it next week if it will!

Please record it if it’s possible. There already is an Obsidian talk channel in Youtube.

(I won’t be able to attend it sadly - I am helping a friend out on that day)

I’m aware of the channel! I still haven’t decided if I want to record the session or not, but in case I don’t, and you can’t make it to the meeting, do not despair! Cat’s wonderful write-up already covers a lot of what I’ll be showing and some more advanced topics (e.g. dataview): Zotero -> zotfile -> mdnotes -> obsidian -> dataview Workflow. Bri’s talk in the Obsidian Office Hours channel could also be of interest to you: Merging library & information science and PKM - How Bri Watson uses Obsidian - YouTube

1 Like

I have a lot of technical (non-academic) PDFs stored locally on my laptop that I want to annotate and move into zotero. Does the workflow differ from PDFs downloaded from the web. When I initially setup Zotero and dragged in a local PDF, I got an error when I tried to extract the annotations with mdnotes. It said that I needed a parent item so I used the create parent item and entered the authors name manually. Once I did that, it was happy and did the md conversion. Can you discuss what else would be considered best practice in respect to manually populating the metadata. Additionally, recommended folder structure in Zotero libary. Thanks!

1 Like

Zotero can extract the metadata of PDFs that have the information properly input and create a parent-item from that. To enable this, go to zotero’s preferences, and in the “General” tab, enable “Automatically retrieve metadata for PDFs”.

For folder structure, you can use zotfile’s wildcards to get it to automatically move your files into particular folders using the wildcards listed here, meaning you can organise your library however you best recall files, whether by collection, year, author’s name, item type, etc. It’s advisable to find what works best for you, rather than go with what others recommewd here. For instance, I recall subjects best, so I have zotero collections organised by subjects, and zotfile automatically moves PDFs into the relevant folders labelled according to the collection (wild card %c). Many people prefer organising by author’s name instead, as that’s how they recall files best - by who wrote it - and therefore can find the article if the PDFs are organised by author’s name.


Edit: I believe I now know, it is the Obsidian Community Talks channel on YouTube. Will be there! Thanks for setting these up!

Hi @miguelmarcos, the talk will be held in Zoom! The announcement (and reminders) for the talks usually happens on Discord (click on the arrow thingy):

The specific message with the announcement is here: Discord

Hope to see you there tomorrow!

Oh, OK. Thanks for the heads up.

Here are the answers to the remaining Q&As from today’s session


This is a great question:

Imagine we are writing a paper and we use citekeys in the [citekey] format for pandoc to render it properly as a citation. but we also have md pages in our vaults which can be found at [[citekey]]

How would I link the md page without having to write these twice when referencing the source in the paper md draft: “extract from paper - [citekey] [[citekey]]”

I’m a bit opinionated on this, my notes use [[citekey]] , and is what I use in obsidian… I don’t use [@citekey] very often because I do my writing in LaTeX, but when I do use them it’s as part of the final product (that I’m exporting to PDF). So far I do it when I bring my paper back from LaTeX → markdown to store in obsidian

When annotations are converted to markdown, is it ok to rename the md file before moving it to Obsidian?

Renaming the file when you create it should be alright! If you rename it afterwards, you might need to update the links added in Zotero by mdnotes

Sometimes when I use Citation plugin Word, usually adds the first & last mame of the author, e.g APA. I just want to automatically display only the last name.

You might need to use a different citation style that uses the authors last names

Do you know any way to get highlights from Calibre/Epubs?

I don’t, there might be a plugin for it that I’m not aware of: plugins [Zotero Documentation]

What is the timeline of this Zotero support/change? Coming soon?

If it’s referring to mdnotes, coming as soon as I have some time! Hard to estimate a timeline

I’m trying to use custom placeholders, which I set in the config editor. They are not being rendered when choosing 'Create full export note". Do you know why?

Probably you are adding the placeholders to the wrong template, the placeholders should be in the Mdnotes Default Template.md file (you’re using single file exports)

what OS are you using? Your screen looks rad!

Thank you! I’m using Ubuntu with i3 https://i3wm.org/

I have Zotero loaded and up, but it looks nothing like what you are working in. What is the tool she is showing us? Vault? Word processor? ???

This is due to my system setup, I’ve added this theme and that changes the Zotero appearance: https://github.com/EliverLara/Nordic (I use a nord color scheme almost everywhere)

how do i get memos to obsidian?

If those are the little yellow notes, then using the Export to markdown menu should do it

My main question is “how do I change how the formatting looks when it gets extracted and plopped into Obsidian” for what it’s worth.

I hope the tag example helped, but if it didn’t let me know, and I can set up a couple of gists with more examples for customizing the formatting


Alright! Videos and handouts are up, and I’ve just added them to the bottom of the main post:

1 Like

I have Zotero loaded and up, but it looks nothing like what you are working in. What is the tool she is showing us? Vault? Word processor? ???

Apparently there are technical reasons well beyond my understanding that limit the Zotero developers’ ability to create a Dark Mode option. Other folkx (also with technical capabilities well beyond mine) created a CSS file to simulate a dark mode. For Windows it is available here; for Mac here. NOTE: The location for Mac is a little tricky: Create a Chrome folder inside /Users/username/Library/Application Support/Zotero/Profiles/bunchofletters.default (mine is named “lio0a6gq.default”) and copy the userChrome.css to that folder.

Thanks for the video! I was surprised that the obsidian citations plugin wasn’t even mentioned. I assume that this is because @argentum isn’t using it? In any case, I’m trying to wrap my head around how mdnotes and citations relate to each other.

There is obviously some overlap in the functionality: both can create notes with metadata for articles/books in zotero. One pushes them from zotero to obsidian, the other pulls them out of zotero to obsidian. Both allow me to customize the templates based on which those notes are created, so I assume, I can create more or less identical notes in both. But I like the way how citations allows me to create a reference to a book/article for which I don’t even have a literature note yet (it creates a link to a non-existing note). I also like the way how I can create a literature note in obsidian and pull over the metadata from zotero.

So what I’m wondering is: does it make sense to use both the citations plugin and the mdnotes add-on? If so, what is a good way of setting this up?

At the moment, I’m thinking to create the meta-data notes with the citations plugin and push my annotations notes from zotero and transclude them in the meta-data note. But I still need to figure out how I need to name my notes-files so that this can be automated via a template. I’m also not sure what mdnotes features I’m missing if I go down that path…

Thanks in advance for any clarifications.

@tophee we do cover it (albeit very very briefly) in the 2nd part when I talk about Better Bibtex. The talk itself was more about how to set up Zotero and its plugins (but it is roughly what I covered for the citations plugin too).

There is some overlap between Mdnotes and the citations plugins, it depends on the information you might want to include in your templates which one might be more suitable.
The citations plugin has access to what is exported in your bib/csl file. Mdnotes has access to any Zotero field and meta information (e.g. collection names, dates, etc.) or the ability to add/format your extracted annotations and fields, e.g. what Cat demoed here and/or adding block IDs to your extracted annotations.

You can still use both, depending on your preferences. Personally, I find the citations plugin great for navigating my literature notes in Obsidian! But as you said, you can use it to create templates too and export annotations with mdnotes or just the citations plugin too!