🤗 HuggingFace Papers Scraper

Track trending AI research from HuggingFace Papers and turn it into clean, structured data for agents, dashboards, and research workflows.

Choose a period (Daily, Weekly, or Monthly), pick an end date, and scrape up to 200 papers with titles, abstracts, summaries, upvotes, PDF links, and more. The actor starts from the end date and walks older papers.

💡 Perfect for…

• 🤖 AI agents: Fresh research context with pdf_url, project_url, and agent_command.
• 📚 RAG pipelines: Index abstracts, summaries, and source URLs with citations.
• 🔬 Research monitoring: Spot emerging models, benchmarks, and methods.
• 📈 Trend analysis: Compare upvotes, organizations, and topics over time.

✨ Why you'll love this scraper

• 📅 Flexible windows: Daily, weekly, or monthly views from a date you choose.
• 📄 Research-ready fields: Abstract, summary, PDF link, project page, and HuggingFace read command.
• 🏢 Attribution: Submitter and organization details when available.

📦 What's inside the data?

• Core: url, title, published_date, submitted_date
• People/org: submitted_by, submitted_by_url, organization, organization_url
• Engagement: upvotes
• Content: abstract, summary
• Links: pdf_url, project_url, agent_command (e.g. hf papers read 2605.29486)

🚀 Quick start

1. Choose period: Daily, Weekly, or Monthly.
2. Optionally set end_date (YYYY-MM-DD). Empty or future → uses today.
3. Set max_papers (up to 200).
4. Start and export JSON, CSV, or Excel.

---

Example input

{
  "period": "Daily",
  "end_date": "2026-06-01",
  "max_papers": 100
}

Example output

{
  "url": "https://huggingface.co/papers/2605.29486",
  "title": "PhoneWorld: Scaling Phone-Use Agent Environments",
  "published_date": "2026-05-28T00:00:00",
  "submitted_date": "2026-05-29T00:00:00",
  "submitted_by": "Zhengyang Tang",
  "submitted_by_url": "https://huggingface.co/tangzhy",
  "organization": "shanghai ailab",
  "organization_url": "https://huggingface.co/ShanghaiAiLab",
  "upvotes": 2,
  "abstract": "PhoneWorld is a pipeline that transforms real GUI trajectories…",
  "summary": "A central bottleneck for phone-use agents…",
  "pdf_url": "https://arxiv.org/pdf/2605.29486",
  "project_url": null,
  "agent_command": "hf papers read 2605.29486"
}

Parameter	Type	Required	Description
period	string	No	Daily, Weekly, or Monthly. Default: Daily.
end_date	string	No	Latest date to start from (YYYY-MM-DD). Empty/future → today.
max_papers	integer	No	How many papers to collect. Min 10, max 200, default 100.