Paper Search MCP

🚀 Paper Search MCP Service

A Model Context Protocol (MCP) server for searching and downloading academic papers from multiple sources, including arXiv, PubMed, bioRxiv, and Sci-Hub (optional). It's designed to seamlessly integrate with large language models like Claude Desktop.

🚀 Quick Start

paper-search-mcp is a Python-based MCP server that allows users to search for and download academic papers from multiple platforms. It offers search tools (e.g., search_arxiv) and PDF download tools (e.g., download_arxiv), making it ideal for researchers and AI-driven workflows. Built with the MCP Python SDK, it integrates smoothly with LLM clients like Claude Desktop.

Installation

paper-search-mcp can be installed using uv or pip. There are two methods below: a quick start for immediate use and a detailed development installation.

Quick Installation with Smithery

Quickly install via Smithery:

# Run the following command in the terminal:
uv add @openags/paper-search-mcp

This will automatically download and install paper-search-mcp along with all its dependencies.

Manual Installation with pip

If you can't use Smithery, you can install it manually:

1. Clone the repository:

   git clone https://github.com/yourusername/paper-search-mcp.git
   cd paper-search-mcp
   

2. Install the dependencies:

   pip install -r requirements.txt
   

3. Install the package:

   python setup.py install
   

Initialization and Usage

After installation, follow these steps to get started quickly:

1. Initialize the MCP service:

   from paper_search_mcp import PaperSearchMCP
   server = PaperSearchMCP()
   

2. Start the server:

   uv run your_project_name
   

3. Search for papers using the API:

   curl http://localhost:5000/api/search?q=quantum+computing
   

4. Download a PDF:

   curl -o paper.pdf http://localhost:5000/api/download/1234
   

Configuration with Claude Desktop

To integrate paper-search-mcp with Claude Desktop, follow these steps:

1. Configure the MCP proxy in Claude:

   proxies:
     paper_search_mcp:
       type: http
       url: http://localhost:5000
   

2. Call the paper search function using Claude:

   from claude import Claude
   claire = Claude(license_key)
   response = claire(
       "Search for the latest papers on quantum computing.",
       tools=[PaperSearchMCP Tool]
   )
   

✨ Features

• Multi-source Support: Search for and download papers from arXiv, PubMed, bioRxiv, and Sci-Hub (optional).
• Standardized Output: Return a consistent dictionary format via the Paper class.
• Asynchronous Tools: Efficiently handle network requests using httpx.
• MCP Integration: Compatible with MCP clients to enhance LLM context.
• Scalable Design: Easily add new academic platforms by extending the academic_platforms module.

📦 Development Installation

Environment Setup

1. Clone the repository:

   git clone https://github.com/yourusername/paper-search-mcp.git
   cd paper-search-mcp
   

2. Create a virtual environment and install the dependencies:

   python -m venv venv
   source venv/bin/activate  # Use `venv\Scripts\activate` on Windows
   pip install -r requirements.txt
   

3. Run the tests:

   pytest tests/
   

🤝 Contribution

Contributions are welcome! Please follow these steps to submit your code:

1. Fork the repository and make changes on your branch.
2. Commit and push the changes.
3. Create a Pull Request to the main branch.

Adding a New Platform

To add a new academic platform, follow these steps:

1. Create a new module in the academic_platforms/ folder.
2. Implement the search and download functions.
3. Update the test cases.
4. Submit to the repository.

📋 To-Do List

Planned Academic Platforms

• [√] arXiv
• [√] PubMed
• [√] bioRxiv
• [√] medRxiv
• [√] Google Scholar
• [ ] Semantic Scholar
• [ ] PubMed Central (PMC)
• [ ] Science Direct
• [ ] Springer Link
• [ ] IEEE Xplore
• [ ] ACM Digital Library
• [ ] Web of Science
• [ ] Scopus
• [ ] JSTOR
• [ ] ResearchGate
• [ ] CORE
• [ ] Microsoft Academic

📄 License

This project is distributed under the MIT License. For more details, see the LICENSE file.

Happy researching with paper-search-mcp! If you encounter any issues, please open an issue on GitHub.