AutoResearch-FactoryAutoResearch-Factory/Agon/skills/arxiv-tools/SKILL.md

arxiv-tools

Skill54 starsChanged 4 months agoMITView on GitHub

This skill should be used when the user asks to "search for papers", "get paper info", "download LaTeX source", "generate BibTeX citation", "find citing papers", or provides an arXiv ID or arXiv URL.

In this repository4 other files it gives its agents.
- arisSkill
- novelty-checkSkill
- server-healthSkill
- sibylSkill

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.

Common
Distinctive
Very specific

Distinctive:
search/cited prefer S2: more complete citation data
line 98

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.

Files

SKILL.md

arXiv Paper Search and Analysis

Six subcommands available: search (find papers), info (metadata), tex (download full text), infotex (info + tex combined), bib (BibTeX), cited (reverse citation lookup).

How to run

Script dependencies are declared in inline metadata at the top of the file — you are responsible for installing them yourself.

Subcommands

search — find papers

uv run "${CLAUDE_PLUGIN_ROOT}/skills/arxiv-tools/arxiv_tool.py" search "keywords"
uv run "${CLAUDE_PLUGIN_ROOT}/skills/arxiv-tools/arxiv_tool.py" search "keywords" --max 10
uv run "${CLAUDE_PLUGIN_ROOT}/skills/arxiv-tools/arxiv_tool.py" search "keywords" --source s2|openalex|arxiv

S2-specific filter parameters:

--year 2024                    # single year
--year 2020-2024               # year range
--fields-of-study "Computer Science,Physics"
--pub-types "JournalArticle,Conference"
--min-citations 50
--venue "NeurIPS"
--open-access                  # open-access only

S2 bulk search (up to 1000 results, supports sorting and pagination):

--bulk --sort "citationCount:desc"
--bulk --token <token from previous page>   # pagination

info — get paper metadata (does not download tex)

uv run "${CLAUDE_PLUGIN_ROOT}/skills/arxiv-tools/arxiv_tool.py" info <arXiv ID>

Prints the metadata (title, authors, date, categories, PDF URL, abstract).

tex — download full paper source

uv run "${CLAUDE_PLUGIN_ROOT}/skills/arxiv-tools/arxiv_tool.py" tex <arXiv ID>

Downloads the tex source and returns the directory path and structure, preserving full LaTeX formatting and figures. If the source is unavailable, falls back to PDF download and text extraction.

infotex — info + tex combined

uv run "${CLAUDE_PLUGIN_ROOT}/skills/arxiv-tools/arxiv_tool.py" infotex <arXiv ID>

Runs info then tex in sequence on a single paper. Use this when you want both the at-a-glance context and the full text in one shot.

bib — generate BibTeX citation

uv run "${CLAUDE_PLUGIN_ROOT}/skills/arxiv-tools/arxiv_tool.py" bib <arXiv ID>
uv run "${CLAUDE_PLUGIN_ROOT}/skills/arxiv-tools/arxiv_tool.py" bib <arXiv ID> -o references.bib  # append to file

cited — reverse citation lookup

Find which papers cite a given paper.

uv run "${CLAUDE_PLUGIN_ROOT}/skills/arxiv-tools/arxiv_tool.py" cited <arXiv ID>
uv run "${CLAUDE_PLUGIN_ROOT}/skills/arxiv-tools/arxiv_tool.py" cited <arXiv ID> --max 50
uv run "${CLAUDE_PLUGIN_ROOT}/skills/arxiv-tools/arxiv_tool.py" cited <arXiv ID> --offset 20         # pagination
uv run "${CLAUDE_PLUGIN_ROOT}/skills/arxiv-tools/arxiv_tool.py" cited <arXiv ID> --source s2|openalex

Data source fallback

Subcommand	Fallback order
search	S2 → OpenAlex → arXiv
cited	S2 → OpenAlex
info / bib	local cache → OpenAlex → S2 → arXiv
tex	local cache → arXiv
infotex	info fallback, then tex fallback (runs sequentially)

search/cited prefer S2: more complete citation data
info/bib prefer OpenAlex: most lenient rate limit (0.1s/req vs S2 2s vs arXiv 5s)

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

It worked
Partly
It didn't

What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free