mosaic notebook

Creates a Google NotebookLM notebook from a search, or from a folder of PDFs, and queues the artifacts you ask for: an audio overview, a video, slides, a quiz and more.

Usage

mosaic notebook create "Hall thrusters" --query "hall thruster anomalous transport" --oa-only
mosaic notebook create "Hall thrusters" --query "hall thruster" -y 2020-2025 --oa-only --podcast --briefing
mosaic notebook create "My library" --from-dir ~/mosaic-papers --slide-deck --mind-map

Behaviour

Needs the notebooklm extra and a one-time notebooklm login. NotebookLM has no official API: the integration uses the unofficial notebooklm-py client, and can break when Google changes the service.

Give exactly one of --query or --from-dir. With --query, mosaic searches, downloads the PDFs and imports them; the search filters (-y, -a, -j, -f, --raw-query) apply only there. --oa-only and --pdf-only apply in both modes.

The artifact flags queue generation in NotebookLM after the import; the artifacts appear in the notebook when Google finishes them.

See also: NotebookLM.

Options

mosaic notebook create

Create a NotebookLM notebook from a search query or a directory of PDFs.

mosaic notebook create [OPTIONS] NAME

Argument	Value	Description
NAME	TEXT	Notebook name

Option	Value	Default	Description
--query, -q	TEXT		Search query to populate the notebook
--from-dir	PATH		Import all PDFs from this directory
--max, -n	INTEGER	10	Max results per source
--oa-only			Only include open-access papers
--pdf-only			Only include papers with a downloadable PDF
--podcast			Queue an Audio Overview after import
--video			Queue a Video Overview after import
--briefing			Queue a Briefing Doc after import
--study-guide			Queue a Study Guide after import
--quiz			Queue a Quiz after import
--flashcards			Queue Flashcards after import
--infographic			Queue an Infographic after import
--slide-deck			Queue a Slide Deck after import
--data-table			Queue a Data Table after import
--mind-map			Queue a Mind Map after import
--year, -y	TEXT		Year filter: "2020", "2020-2024", or "2020,2022,2024"
--author, -a	TEXT		Author name filter (repeatable)
--journal, -j	TEXT		Journal name filter (substring match)
--field, -f	TEXT	all	Scope query to "title", "abstract", or "all" (default)
--raw-query	TEXT		Raw query sent directly to source APIs, bypassing all field transforms
--download-dir	TEXT		Override PDF download directory for this run