research

name	research
description	Search, fetch, download, and extract ML/AI research papers from 7 free academic sources. Find and download ML datasets from 5 free sources (HuggingFace, OpenML, UCI, Papers with Code, Kaggle). Review a paper, critique methodology, assess reproducibility, evaluate experimental design. Convert research papers, articles, or technical documents into working code prototypes. Use when the user wants to find papers, search arxiv, get citations, download a PDF, extract text from a paper, find/download datasets, review/critique a research paper, implement a paper, prototype an algorithm, or convert research to working code.
allowed-tools	Bash(uv run * scripts/scientific_search.py *) Bash(uv run * scripts/search.py *) Bash(uv run * scripts/fetch.py *) Bash(uv run * scripts/download.py *) Bash(uv run * scripts/extract.py *) Bash(uv run * scripts/analyze_document.py *) Bash(uv run * scripts/datasets.py *) Read Write WebFetch Glob Grep
argument-hint	search query, paper ID, or path/URL to implement (e.g. "transformer attention", "2401.12345", or "papers/attention.pdf")
model	sonnet
effort	medium
compatibility	>=1.0
metadata	{"category":"research","tags":["arxiv","semantic-scholar","papers-with-code","datasets","huggingface","openml","kaggle","paper-review","prototype"],"phase":"research"}

Academic Paper Research & Dataset Discovery

Instructions and tools for searching, fetching, and extracting ML/AI research papers, plus finding and downloading datasets.

Prerequisites

which pdftotext || echo "MISSING: brew install poppler (macOS) or apt install poppler-utils (Linux)"

Available sources (7 total)

Source	Search	Fetch	Best for
arXiv	yes	yes	ML/AI preprints
Semantic Scholar	yes	yes	Citations, open-access PDFs
Papers with Code	yes	yes	Papers linked to GitHub repos
Hugging Face	yes	via arXiv	Trending daily papers
JMLR	yes	yes	Peer-reviewed ML journal
ACL Anthology	no	by ID	NLP conference papers
OpenScholar	no	no	Q&A synthesis over 45M papers (URL only)

Commands

Search

uv run ${CLAUDE_SKILL_DIR}/scripts/search.py "$ARGUMENTS" --source arxiv --max 5

Flags: --source (arxiv, semantic_scholar, papers_with_code, huggingface, jmlr, openscholar), --max N, --cat cs.AI,cs.LG, --sort relevance|date

Fetch metadata

uv run ${CLAUDE_SKILL_DIR}/scripts/fetch.py <paper_id>

Auto-detects source: 2401.12345 (arXiv), 2022.acl-long.220 (ACL), v22/19-920 (JMLR), 40-char hex (Semantic Scholar)

Download PDF

uv run ${CLAUDE_SKILL_DIR}/scripts/download.py <paper_id> -o ./papers/

Extract text