Hugging Face Papers

---
name: huggingface-papers
description: Look up and read Hugging Face paper pages as markdown and use the papers API for structured metadata (authors, linked models/datasets/spaces, GitHub repo, project page). Use when the user shares an HF paper URL, an arXiv URL or ID, or asks to summarize or analyze an AI paper.
---

# Hugging Face Paper Pages

hf.co/papers is built on top of arXiv for AI/CS research, with daily-papers feeds, upvotes, and links to model checkpoints, datasets, and Spaces.

## Workflow
1. Parse the paper ID (arXiv ID) from whatever the user provides - HF URL, arXiv abs/pdf URL, or a bare ID like `2602.08025`.
2. Fetch the paper as markdown: `curl -s "https://huggingface.co/papers/{ID}.md"` (falls back to the HF page HTML when no arXiv HTML exists).
3. Fetch structured metadata as JSON: `curl -s "https://huggingface.co/api/papers/{ID}"` for authors and linked models/datasets/spaces, GitHub, and project page.
4. Handle a 404 as "not yet indexed on hf.co/papers" and follow the skill's error-handling guidance.
5. Summarize or analyze using the fetched content and metadata.

Full skill & source: https://github.com/huggingface/skills/tree/ea9a24f38a98baf2430dab3bcb31a89753ae9e6d/skills/huggingface-papers