https://github.com/ultimatile/arxiv-skills
https://github.com/ultimatile/arxiv-skills
agent-skills
Last synced: 20 days ago
JSON representation
- Host: GitHub
- URL: https://github.com/ultimatile/arxiv-skills
- Owner: ultimatile
- License: mit
- Created: 2025-12-08T05:11:07.000Z (6 months ago)
- Default Branch: main
- Last Pushed: 2026-04-30T22:13:42.000Z (30 days ago)
- Last Synced: 2026-05-10T22:54:44.826Z (20 days ago)
- Topics: agent-skills
- Language: Python
- Homepage:
- Size: 81.1 KB
- Stars: 19
- Watchers: 0
- Forks: 2
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# arXiv Skills Repository
Custom Claude skills for working with arXiv papers - fetching, searching, and converting to reference documentation.
## Available Skills
### 1. arxiv-doc-builder
Automatically convert arXiv papers to well-structured Markdown documentation for implementation reference.
**Description:** Automatically fetches arXiv papers (LaTeX source or PDF), converts them to Markdown, and generates implementation-ready reference documentation with preserved mathematics and section structure.
**Use when:** You need Claude to convert an arXiv paper into readable Markdown documentation for code implementation or research reference.
**Capabilities:**
- Automatic paper fetching with source→PDF fallback
- LaTeX source → Markdown conversion (via pandoc)
- PDF → Markdown text extraction
- Mathematical formula preservation in MathJax/LaTeX format
- Section structure and hierarchy preservation
- Advanced vision-based PDF conversion available for complex formulas
### 2. arxiv-lookup
Lightweight scripts for querying the arXiv API directly — get journal DOIs from arXiv IDs, or search for papers by title/keyword.
**Description:** Look up arXiv paper metadata via the arXiv API. Get journal DOIs from arXiv IDs (for OpenAlex integration), or find arXiv IDs from title/keyword search (for arxiv-doc-builder).
**Use when:** You need Claude to look up a paper's journal DOI, or find arXiv IDs by searching titles, authors, or categories.
**Capabilities:**
- Get journal DOI from an arXiv ID (`get_doi.py`)
- Search arXiv by title, author, abstract, or category (`search_id.py`)
- Supports arXiv API field prefixes (`ti:`, `au:`, `abs:`, `cat:`) and boolean operators
## Installation
```bash
claude plugin marketplace add ultimatile/arxiv-skills
claude plugin install arxiv-skills
```
### Install `convert-paper` as a Global CLI Tool
`arxiv-doc-builder` can also be installed as a standalone CLI tool to run `convert-paper` from anywhere:
```bash
uv tool install --from 'arxiv-doc-builder @ git+https://github.com/ultimatile/arxiv-skills.git#subdirectory=skills/arxiv-doc-builder'
```
```bash
# Usage
convert-paper 2409.03108
convert-paper 2409.03108 --output-dir ~/papers
```
**Requirements:**
- Python 3.8+
- pandoc (for LaTeX conversion: `brew install pandoc`)
- poppler-utils (for advanced PDF processing: `brew install poppler`)
- Python dependencies auto-installed via uv