Paper Search

v1.14.0Search & Data Extractionstable

Enables searching and downloading academic papers from multiple sources including arXiv, PubMed, bioRxiv, Google Scholar, and Semantic Scholar. Provides standardized tools for research workflows and OpenAI Deep Research integration.

ai-scientistarxiv-papersmcp-serverpaper-searcharxiv
Share:
1,524
Stars
0
Downloads
0
Weekly
0/5

What is Paper Search?

Paper Search is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to enables searching and downloading academic papers from multiple sources including arxiv, pubmed, biorxiv, google scholar, and semantic scholar. provides standardized tools for research workflows and o...

Enables searching and downloading academic papers from multiple sources including arXiv, PubMed, bioRxiv, Google Scholar, and Semantic Scholar. Provides standardized tools for research workflows and OpenAI Deep Research integration.

This server falls under the Search & Data Extraction category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Enables searching and downloading academic papers from multi

Use Cases

Search and download academic papers from arXiv, PubMed, and bioRxiv.
Integrate with OpenAI Deep Research for enhanced research workflows.
Standardize paper retrieval across multiple academic sources.
openags

Maintainer

LicenseMIT License
Languagepython
Versionv1.14.0
UpdatedMay 22, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx paper-search-mcp

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Paper Search

Paper Search MCP provides a unified interface for searching and downloading academic papers across 24 sources including arXiv, PubMed, bioRxiv, medRxiv, Semantic Scholar, Google Scholar, CORE, Zenodo, IEEE Xplore, and ACM Digital Library. It exposes a high-level search_papers tool that queries multiple platforms concurrently and deduplicates results, plus a download_with_fallback tool that tries multiple open-access routes before optionally falling back to Sci-Hub. Researchers and AI scientists use it to power literature review workflows, integrate with OpenAI Deep Research pipelines, and retrieve papers without manually visiting each database.

Prerequisites

  • Python 3.10 or later with uv or pip available
  • An MCP client such as Claude Desktop
  • Optional: Free API keys for Semantic Scholar (PAPER_SEARCH_MCP_SEMANTIC_SCHOLAR_API_KEY), CORE (PAPER_SEARCH_MCP_CORE_API_KEY), IEEE Xplore (PAPER_SEARCH_MCP_IEEE_API_KEY), and ACM DL (PAPER_SEARCH_MCP_ACM_API_KEY) to increase rate limits
  • Optional: An email address for Unpaywall OA lookup (PAPER_SEARCH_MCP_UNPAYWALL_EMAIL)
1

Install the package

Install paper-search-mcp using uvx (recommended) or pip. The uvx method runs it in an isolated environment without polluting your global Python installation.

# Recommended
uvx paper-search-mcp

# Or with pip
pip install paper-search-mcp
2

Obtain optional API keys

Most sources work without API keys, but free keys for Semantic Scholar, CORE, and IEEE Xplore improve rate limits and unlock additional platforms. Register at each provider's developer portal.

3

Add the server to your MCP client configuration

Edit claude_desktop_config.json to add paper-search-mcp with any optional API keys as environment variables.

{
  "mcpServers": {
    "paper-search": {
      "command": "uvx",
      "args": ["paper-search-mcp"],
      "env": {
        "PAPER_SEARCH_MCP_UNPAYWALL_EMAIL": "[email protected]",
        "PAPER_SEARCH_MCP_SEMANTIC_SCHOLAR_API_KEY": "your-s2-key",
        "PAPER_SEARCH_MCP_CORE_API_KEY": "your-core-key",
        "PAPER_SEARCH_MCP_IEEE_API_KEY": "your-ieee-key",
        "PAPER_SEARCH_MCP_GOOGLE_SCHOLAR_PROXY_URL": "http://your-proxy:port"
      }
    }
  }
}
4

Restart your MCP client

Quit and reopen Claude Desktop so it connects to the newly configured paper-search server.

5

Search for papers

Ask Claude to find papers on a topic. The search_papers tool queries all enabled sources concurrently and returns deduplicated results ranked by relevance.

Paper Search Examples

Client configuration

claude_desktop_config.json entry for paper-search-mcp with optional API keys.

{
  "mcpServers": {
    "paper-search": {
      "command": "uvx",
      "args": ["paper-search-mcp"],
      "env": {
        "PAPER_SEARCH_MCP_UNPAYWALL_EMAIL": "[email protected]",
        "PAPER_SEARCH_MCP_SEMANTIC_SCHOLAR_API_KEY": "your-s2-api-key",
        "PAPER_SEARCH_MCP_CORE_API_KEY": "your-core-api-key",
        "PAPER_SEARCH_MCP_IEEE_API_KEY": "your-ieee-api-key"
      }
    }
  }
}

Prompts to try

Research queries you can ask Claude once the paper-search MCP server is connected.

- "Search for papers on transformer attention mechanisms from the last 2 years"
- "Find recent arXiv preprints on diffusion models for protein structure prediction"
- "Search PubMed for clinical trials on GLP-1 receptor agonists and weight loss"
- "Download the paper arxiv:2310.06825 and summarize its key contributions"
- "Find papers on CRISPR base editing from bioRxiv and compare their approaches"

Troubleshooting Paper Search

Google Scholar searches return no results or fail with bot-detection errors

Google Scholar aggressively blocks automated access. Set PAPER_SEARCH_MCP_GOOGLE_SCHOLAR_PROXY_URL to a rotating residential proxy URL to bypass bot detection, or rely on the other 23 sources which do not have this restriction.

Semantic Scholar or CORE searches are rate-limited

Register for a free API key at api.semanticscholar.org and api.core.ac.uk, then set PAPER_SEARCH_MCP_SEMANTIC_SCHOLAR_API_KEY and PAPER_SEARCH_MCP_CORE_API_KEY in your environment configuration.

Unpaywall download fallback does not work

Unpaywall requires a valid email address to use its API. Set PAPER_SEARCH_MCP_UNPAYWALL_EMAIL to your institutional or personal email. Without this variable, Unpaywall is disabled as a fallback source.

Frequently Asked Questions about Paper Search

What is Paper Search?

Paper Search is a Model Context Protocol (MCP) server that enables searching and downloading academic papers from multiple sources including arxiv, pubmed, biorxiv, google scholar, and semantic scholar. provides standardized tools for research workflows and openai deep research integration. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Paper Search?

Follow the installation instructions on the Paper Search GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with Paper Search?

Paper Search works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Paper Search free to use?

Yes, Paper Search is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.

Browse More Search & Data Extraction MCP Servers

Explore all search & data extraction servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "paper-search-mcp": { "command": "npx", "args": ["-y", "paper-search-mcp"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Paper Search?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides