MCP Catalogs
Homepaper-search-mcp screenshot

paper-search-mcp

by openags·1,461·Score 56

MCP server for searching and downloading academic papers from multiple sources with free-first strategy.

ai-llmdeveloper-toolsresearch
153
Forks
36
Open issues
1 mo ago
Last commit
2d ago
Indexed

Overview

paper-search-mcp is a Python-based MCP server that provides tools for searching, downloading, and extracting academic papers from various platforms. It features a two-layer architecture with unified tooling for multi-source search and platform-specific connectors. The server follows a free-first strategy, prioritizing public and open data sources while supporting optional API keys for enhanced functionality. It supports a wide range of academic platforms including arXiv, PubMed, bioRxiv, and more, with standardized output and extensible design.

Try asking AI

After installing, here are 6 things you can ask your AI assistant:

you:Researchers can quickly find and download papers from multiple sources in one search.
you:AI systems can access academic papers as context for enhanced responses.
you:Students and academics can build paper collection workflows with automated downloading and organization.
you:Is this server compatible with Claude Desktop?
you:Do I need API keys to use this server?
you:Does it support Sci-Hub for downloading papers?

When to choose this

Choose Paper Search MCP when you need comprehensive academic paper search capabilities across multiple repositories with a free-first approach and standardized output formats for AI integration.

When NOT to choose this

Don't choose this if you need access to paid-only journals (IEEE Xplore, ACM DL require API keys and are still in skeleton stage) or if your institution restricts access to many of the supported open repositories.

Tools this server exposes

12 tools extracted from the README
  • search_papers

    Search for academic papers across multiple sources with deduplication

  • download_with_fallback

    Download papers using open access links with fallback chain

  • search_arxiv

    Search for papers specifically on arXiv

  • search_pubmed

    Search for biomedical papers on PubMed

  • search_semantic_scholar

    Search papers using Semantic Scholar API

  • search_crossref

    Search papers using Crossref metadata API

  • search_pubmed_central

    Search for open access papers in PubMed Central

  • search_google_scholar

    Search Google Scholar for academic papers

  • search_bioRxiv

    Search for preprint papers on bioRxiv

  • search_core

    Search for papers in the CORE aggregation service

  • search_unpaywall

    Look up open access versions of papers using DOI

  • read_paper

    Extract text content from a downloaded paper

Comparable tools

semantic-scholar-mcparxiv-search-mcppubmed-mcpgoogle-scholar-mcp

Installation

Method 1 — Smithery (Recommended)

smithery install @openags/paper-search-mcp

Method 2 — uvx

uvx paper-search-mcp

Method 3 — uv

uv add paper-search-mcp

Method 4 — pip

pip install paper-search-mcp

Claude Desktop Configuration

Add to Claude Desktop config:

{
  "mcpServers": {
    "paper-search": {
      "command": "uvx",
      "args": ["paper-search-mcp", "stdio"]
    }
  }
}

FAQ

Is this server compatible with Claude Desktop?
Yes, it works with Claude Desktop and other MCP clients. The README includes specific configuration instructions.
Do I need API keys to use this server?
Most functionality works without API keys, but optional keys can improve rate limits and access for certain platforms like Semantic Scholar.
Does it support Sci-Hub for downloading papers?
Sci-Hub is available as an optional fallback, but it's unstable and user responsibility. The server prioritizes legal open access sources first.

Compare paper-search-mcp with

GitHub →

Last updated · Auto-generated from public README + GitHub signals.