Library Management

Beaver can help you browse, search, and organize your Zotero library. It can find items by specific field conditions, explore collections and tags, and organizing items in bulk.

Organizing New Additions

As your library grows, it's easy for new papers to pile up without proper organization. Beaver can help you stay on top of recent additions:

I added several papers this week. Organize them into appropriate collections based on their topics.
Find all unfiled papers and suggest which collections they belong in.
Tag my recent additions as "to-read" so I remember to review them.

Project-Based Organization

When starting a new project or paper, Beaver can help you gather and organize relevant references:

I'm starting a literature review on ____. Create a collection and add all relevant papers from my library.
Find all papers in my library that discuss both machine learning and healthcare, and add them to my "ML Healthcare" collection.

Batch Jobs

When a request covers more items than a single turn can handle, Beaver runs it as a batch job. It builds a list of every item in scope and works through them one at a time, either changing an item or recording that it needs no change. Beaver switches to a batch job automatically for large-scale sorting, tagging, metadata editing, annotating, and extraction, and you can also ask for one directly.

Sort all unfiled items into my existing collections by subject. Don't create new ones.
Add consistent subject tags to every untagged item in my library.
Find every item missing a DOI or journal and fix what you can look up.

How a Batch Job Runs

  1. Approval: Nothing runs until you approve it. The approval card shows the exact item count, the goal of the job, the maximum credit cost, and a warning about any destructive library changes.
  2. Progress: A live progress bar shows the count so far and a running tally of where items are landing. An item counts as done only when the work actually happened.
  3. Receipt: When the last item has a decision, the job closes with a receipt: how many items went where, how many needed no change and why, and what could not be read.

Limits

Limits are per batch job. Multiple batch jobs can run in a single request or across several requests.

WorkItems per job
Sort items into collections1,000
Tag items1,000
Fix metadata1,000
Read documents and extract information180
Annotate documents100
Write notes100

Getting Good Results

  • Name the population clearly. "All unfiled items" leaves less room for guessing than "a few of my unfiled items".
  • Say what a good outcome looks like for one item. The goal is judged one item at a time. "File it into the existing subject collection that matches what the paper is about" gives Beaver something it can hold a single item against; "tidy up my library" does not.
  • Define your tags and collections, not just name them. Beaver derives what a tag or collection means from its name and contents. A one-line definition, for example in your custom prompt, makes results far more consistent across hundreds of items.
  • One kind of work at a time. Tagging, sorting, and writing notes are three different kinds of work. Use three separate requests so each has a clearly defined goal.

For benchmarks and more examples, see the batch jobs overview.

Automating Common Tasks

If you frequently perform the same organization tasks, you can save time by creating actions. Actions are reusable prompts you define once and trigger anytime with a keyboard shortcut or the slash menu.

For example, you might create actions for:

  • "Organize all unfiled items from the last week into appropriate collections"
  • "Review item metadata for all items added this week"
  • "Add the citation count to the extra field for all items that don't have it"

See the Actions documentation for setup instructions and more examples.

You Stay in Control

When Beaver modifies your library (adding tags, creating collections, moving items), all changes require your approval by default. Beaver will show you exactly what it plans to do, and nothing happens until you confirm. This ensures you always have the final say over your library organization.

You can configure this behavior in Beaver settings:

  • Always ask (default): Review every proposed change before it's applied
  • Auto-apply: Trust Beaver to make changes automatically
  • Continue without applying: Let Beaver propose changes that you can review and apply later

Large Tasks

Small organization tasks run directly in the conversation. Once a request covers more items than a single turn can handle, Beaver runs it as a batch job instead, so large-scale sorting, tagging, and metadata fixes finish rather than stopping partway. There is no need to split a large task into smaller ones yourself.