arxiv_search

MCP tool from Arxiv MCP Server by cyanheads

Search arXiv papers by query with category and sort filters.

How to use it

arxiv_search is exposed by the Arxiv MCP Server MCP server. Add the server to your MCP client (Claude Desktop, Cursor, Windsurf and others), and the arxiv_search tool becomes available to the model automatically. See the full listing for setup details and every tool this server provides.

Install Arxiv MCP Server

$docker run -p 3010:3010 arxiv-mcp-server

FULL ARXIV MCP SERVER LISTING

Other tools in Arxiv MCP Server (5)

arxiv_get_metadata

Get full metadata for one or more arXiv papers by ID.

arxiv_list_categories

List arXiv category taxonomy, optionally filtered by group.

arxiv_read_paper

Fetch the full text content of an arXiv paper from its HTML rendering, or from the PDF when no render exists.

max_characters

defaults to 100,000; pass null for the whole paper in one call. Raw HTML can be 500KB-3MB+ for math-heavy papers, which is more than most clients accept in a single tool result — page with start instead

submitted_from

/ submittedto bound the submission date (inclusive, UTC YYYY-MM-DD). Consecutive windows cover the matches with no gap — a paper submitted exactly at a midnight seam falls in both, so de-duplicate by ID — which is how to reach results past the 10,000 pagination ceiling