batch-process

Batch Process Papers

Process multiple papers from a text file containing arXiv IDs or URLs.

Arguments

Parse from: $ARGUMENTS

• FILE: Path to a text file containing arXiv IDs/URLs (one per line)

Options:

• --audio: Generate audio overviews
• --slides: Generate slide decks (default: true)
• --skip-existing: Skip papers that already have notebooks
• --prefix "PREFIX": Add prefix to notebook names

Input File Format

The input file should contain one arXiv ID or URL per line:

2601.02151
2601.03509
https://arxiv.org/abs/2601.03872
2601.03986
https://arxiv.org/pdf/2601.04151.pdf

Lines starting with # are treated as comments and ignored.

Workflow

1. Read the input file
2. Parse and validate all arXiv IDs
3. Create a todo list with all papers
4. For each paper: a. Create NotebookLM notebook b. Add arXiv PDF as source c. Generate slide deck (if --slides) d. Generate audio overview (if --audio) e. Update progress
5. Generate summary report

Example Usage

/notebooklm-papers:batch-process ./my-papers.txt
/notebooklm-papers:batch-process ./reading-list.txt --audio
/notebooklm-papers:batch-process ./papers.txt --prefix "Project X -"

Output

Creates a summary report with:

• List of all processed papers
• Links to created notebooks
• Status of slide deck and audio generation
• Any errors encountered

Error Handling

• Invalid arXiv IDs are logged and skipped
• Network errors trigger one retry
• Generation timeouts are logged and processing continues
• Final report includes all successes and failures