Zotero plugin

Renota for Zotero

Read whole books with an assistant that has read the same pages you have.

Renota lives inside the Zotero reader. Select a passage in an EPUB, a PDF or a web snapshot and get an explanation grounded in this book — the earlier chapters, your own highlights, and nothing you have not read yet.

Why it stands out

No spoilers

Only text before your reading position reaches the model, including search results and reading notes. Read a mystery without ruining it.

In-book search

Every book is chunked and indexed locally, so questions like who a character is can be answered from earlier passages with citations.

Writes back into Zotero

Save an explanation as a highlight comment, append it to a reading note, or keep it as an insight in your knowledge base.

What it does

Selection popup in the reader

Understand, analyze, challenge, translate, who/what, save, ask — right below Zotero's own selection popup.

Recap and chapter summary

Get a reminder of what happened so far, or a summary of the chapter you just finished, built from text before your position.

Your annotations become knowledge

Highlights made in Zotero are mirrored into the knowledge base as they are created, so they can be searched and cited later.

Reading notes

Answers can be appended to a note per book, giving you one growing document instead of scattered annotations.

Works on EPUB, PDF and snapshots

Chapter paths and reading position are tracked where the reader exposes them; scanned PDFs fall back to whatever text layer exists.

Shares one knowledge base

The same clips are visible in the Chrome extension and the Android app when sync is enabled.

How to use it

  1. Install the plugin and set a model

    Install the .xpi, then fill in the provider, base URL, key and model in Zotero's settings under Renota.

  2. Open a book and select text

    The book is indexed in the background. Select a passage to get the action buttons.

  3. Write it back

    Send the answer to a highlight comment or your reading note, or keep it as an insight.

What leaves your machine

Only the text needed to answer the action you started, and never text after your reading position.

  • The book index is built and stored locally in a plugin-owned SQLite file.
  • Your API key lives in Zotero's preferences for this plugin.
  • Sync, if enabled, uploads knowledge entries only — never the book file.

Get the plugin

Download the .xpi from the releases page, then install it from Zotero's plugin manager.

Download .xpi

Latest release

Get it

Build from source

npm run build produces the .xpi

View on GitHub

Requirements: Zotero 7 or newer. An API key from any OpenAI-compatible provider, or Anthropic.

Zotero plugin questions

Which Zotero versions are supported?

Zotero 7 is the target, with compatibility declared for 8, 9 and 10.

Does it send the whole book to the model?

No. Only the selection, a little surrounding context and a handful of retrieved passages are sent — all from before your reading position.

Can I turn off annotation mirroring?

Yes, in settings. Turning it off also removes annotations that were already mirrored into the knowledge base.

Does it work with scanned PDFs?

Only to the extent the file has a text layer. Scanned pages have nothing to select, so use the Android app's OCR for those.

Can the Chrome extension read this library?

Yes, on the same machine: enable the local sync endpoint in the plugin settings and pair the extension with a token. Or use cloud sync.