fetch-papers

Fetch Papers

Fetches and displays the top papers from Hugging Face Daily Papers for a given date.

Arguments

Parse from: $ARGUMENTS

• DATE: Date in YYYY-MM-DD format (defaults to today)
• NUM_PAPERS: Number of papers to fetch (defaults to 10, max 50)

Options:

• --json: Output as JSON format
• --markdown: Output as markdown table (default)
• --save FILE: Save output to file

Workflow

1. Get browser context using mcp__claude-in-chrome__tabs_context_mcp
2. Navigate to https://huggingface.co/papers?date={DATE}
3. Read the page content using mcp__claude-in-chrome__read_page or mcp__claude-in-chrome__get_page_text
4. Extract paper information:
   • Title
   • arXiv ID
   • Authors (if available)
   • Upvote count
   • Abstract preview (if available)
5. Format and display the results

Output Format (Markdown)

# Hugging Face Daily Papers - {DATE}

| # | Title | arXiv | Upvotes | PDF |
|---|-------|-------|---------|-----|
| 1 | {title} | [{id}](https://arxiv.org/abs/{id}) | {votes} | [PDF](https://arxiv.org/pdf/{id}.pdf) |
...

Output Format (JSON)

{
  "date": "2026-01-09",
  "papers": [
    {
      "rank": 1,
      "title": "Paper Title",
      "arxiv_id": "2601.02151",
      "upvotes": 42,
      "arxiv_url": "https://arxiv.org/abs/2601.02151",
      "pdf_url": "https://arxiv.org/pdf/2601.02151.pdf"
    }
  ]
}

Example Usage

/notebooklm-papers:fetch-papers
/notebooklm-papers:fetch-papers 2026-01-08 20
/notebooklm-papers:fetch-papers --json --save papers.json