MCP Research Server

A Model Context Protocol (MCP) server that provides tools for searching and managing research papers from arXiv.

Features

• Search Papers: Search for research papers on arXiv by topic
• Extract Paper Info: Retrieve detailed information about specific papers
• Local Storage: Automatically saves paper information to local JSON files organized by topic

Prerequisites

• Python 3.13 or higher
• uv package manager (recommended) or pip

Installation

1. Clone or navigate to the project directory:

   cd mcp_project
   

2. Create a virtual environment:

   uv venv
   source .venv/bin/activate  # On macOS/Linux
   # or
   .venv\Scripts\activate     # On Windows
   

3. Install dependencies:

   uv pip install arxiv
   

Running the Server

Method 1: Using MCP Inspector (Recommended for Testing)

1. Start the server with MCP Inspector:

   npx @modelcontextprotocol/inspector uv run research_server.py
   

2. Access the Inspector:

   • Open your browser and go to: http://127.0.0.1:6274
   • Use the provided session token for authentication

Method 2: Direct Execution

1. Run the server directly:

   uv run research_server.py
   

Available Tools

1. search_papers

Searches for papers on arXiv based on a topic and stores their information locally.

Parameters:

• topic (str): The topic to search for
• max_results (int, optional): Maximum number of results (default: 5)

Returns: List of paper IDs found

Example:

{
  "topic": "machine learning",
  "max_results": 10
}

2. extract_info

Retrieves detailed information about a specific paper from local storage.

Parameters:

• paper_id (str): The ID of the paper to look for

Returns: JSON string with paper information

Example:

{
  "paper_id": "2301.12345"
}

Project Structure

mcp_project/
├── research_server.py    # Main MCP server implementation
├── pyproject.toml        # Project configuration
├── uv.lock              # Dependency lock file
├── papers/              # Directory where paper data is stored
│   └── [topic_name]/    # Organized by topic
│       └── papers_info.json
└── README.md           # This file

How It Works

1. Paper Search: When you search for papers, the server:

   • Queries arXiv using the provided topic
   • Downloads paper metadata (title, authors, summary, PDF URL, publication date)
   • Saves the information to a JSON file organized by topic
   • Returns the paper IDs for reference

2. Paper Retrieval: When you extract paper info, the server:

   • Searches through all topic directories
   • Finds the requested paper by ID
   • Returns the stored information in JSON format

Data Storage

Paper information is automatically saved to the papers/ directory, organized by topic. Each topic gets its own subdirectory containing a papers_info.json file with all the papers found for that topic.

Troubleshooting

• Python Version: Ensure you're using Python 3.13 or higher
• Dependencies: Make sure all dependencies are installed with uv pip install arxiv
• Virtual Environment: Always activate the virtual environment before running the server
• Permissions: Ensure you have write permissions in the project directory for creating the papers/ folder

Development

To modify or extend the server:

1. Edit research_server.py to add new tools or modify existing ones
2. Use the MCP Inspector to test your changes
3. The server uses FastMCP for easy tool definition and management

License

This project is open source. Feel free to modify and distribute as needed.