get_paper_abstract
Get just the abstract of an arXiv paper. Use this for a quick preview when the user hasn't read the paper yet, not when they provide an arXiv ID to discuss a paper.
LABSCO SUMMARY
Equations survive the trip into the model, because the source is what gets fetched.
Taking the LaTeX rather than a rendered page is the entire reason to choose this over a general paper reader: the expressions arrive as written, which is what makes a derivation discussable. For long papers the section path keeps the cost proportional to what is actually being read — list the headings, fetch the one section, leave the rest on arXiv.
What it is
A server that fetches arXiv papers as LaTeX source rather than rendered output, whole or in parts.
What you get
- The full LaTeX source of a paper by arXiv id, for precise reading of its mathematical expressions
- The abstract on its own, for the case where the paper has not been read yet and may not be the right one
- Section headings listed, then one section pulled by its path when the whole paper will not fit the context
Requirements
An arXiv id, and uv with a Python for the server to run under — UV_PYTHON picks the interpreter. No account and no key.
Setup effort
One command — uvx arxiv-latex-mcp
USE CASES
01
Working through a derivation with the model
Fetch the paper as LaTeX so the expressions arrive as the author typed them, which is what makes the steps discussable rather than approximated.
02
Deciding whether the paper is the one you want
Take the abstract on its own before committing any context to the paper itself.
03
Pulling one section out of a paper too long to load
List the headings, choose the section that matters, and fetch it by its path instead of the whole document.
4 tools
Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.
get_paper_prompt
Recommended default: fetch the full LaTeX source of an arXiv paper for precise interpretation of mathematical expressions.
get_paper_section
Get a specific section of an arXiv paper by section path. Use when the full paper is too long for context or the user wants to focus on a particular section. Use list_paper_sections first to find available paths.
list_paper_sections
List section headings of an arXiv paper. Useful when the full paper is too long for context and you need to identify which sections to fetch individually.
BEFORE YOU INSTALL
What to know first
Tools you get
4 tools when we connected.
Requires
uvx on PATH
CONTEXT COST
Light Standing cost
Rated by the size of the 4 tool definitions that ride along on every turn, not by how many there are
INSTALLATION
Connect your client
Add the block to your client, or run the command directly.
Server setup
Configuration file
JSON
{
"mcpServers": {
"arxiv-latex-mcp": {
"command": "uvx",
"args": [
"arxiv-latex-mcp"
]
}
}
}uvx
TERMINAL
uvx arxiv-latex-mcp
