ArXiv Papers

You can empower your agents to search, analyze, and explore arXiv papers autonomously by using this MCP server. It provides advanced search, author and category discovery, version tracking, citation extraction, related paper discovery, and bibliography export to help you build research workflows and generate insights from ArXiv content.

How to use

You connect to the MCP server with your MCP client and start issuing search and analysis prompts. Use multi-field searches with date ranges to locate papers, retrieve full metadata and versions for a given paper, extract citations, and discover related works. You can also export bibliographies in BibTeX or JSON formats for citations in your manuscripts. Leverage AI-powered prompts to generate literature reviews, identify gaps, compare papers, and analyze trends.

How to install

Prerequisites you need before starting: Node.js (version 14 or newer) and npm (comes with Node.js). Ensure you can run basic Node commands in your shell.

Install and start the MCP server using the provided command that leverages an MCP URL via npx.

{
  "search-papers": {
    "command": "npx",
    "args": ["mcp-remote", "https://search-papers.com/sse"]
  }
}

Additional setup details

You may also configure another transport that uses an SSE endpoint by registering a server with a type and URL when supported by your client. For example, you can point to the SSE endpoint exposed at https://search-papers.com/sse to enable streaming responses from the MCP server.

Configuration examples

{
  "search-papers": {
    "type": "http",
    "name": "search-papers",
    "url": "https://search-papers.com/sse",
    "args": []
  }
}

Usage with a local client

If your client supports local or stdio-based MCP servers, you can run a process that starts the MCP server locally using the same command pattern shown above. The key is to point the client to the server’s transport endpoint and provide an empty argument set if none are needed.

Security and access notes

Only connect to MCP endpoints you trust. Do not expose sensitive endpoints publicly and validate that the endpoints you use support secure transport (HTTPS) and proper authentication if provided by the MCP you are connecting to.

Exporting and tools overview

Use the built-in export features to obtain BibTeX or JSON bibliographies for papers you select. You can also run AI-powered prompts to generate literature reviews, identify gaps, compare papers, and analyze research trends.

Available tools

advanced_search

Perform multi-field searches with date ranges and sorting to locate relevant papers.

search_by_author

Find papers authored by specific individuals with optional affiliation tracking.

search_by_category

Browse papers by ArXiv categories to explore subject areas.

search_by_date_range

Query papers published within a defined time window.

get_paper_by_id

Retrieve complete information for a specific paper by its identifier.

get_paper_versions

Track and display all versions of a paper to observe its evolution.

get_citations

Extract citations from abstracts and metadata for a given paper.

get_related_papers

Discover papers related to a target paper based on content similarity.

export_bibliography

Export selected papers as BibTeX or JSON for bibliographies.

literature_review

Generate comprehensive literature reviews from a set of papers.

research_gaps

Identify gaps and opportunities in the current research landscape.

paper_comparison

Compare multiple papers on methodology, findings, or approaches.

trend_analysis

Analyze research trends over time across topics or categories.

categories_list

List all ArXiv categories for browsing.

trending_papers

Retrieve trending papers by category and time period.