ArXiv论文搜索MCP服务器

mcp-arxiv-research-server

A custom MCP (Model Context Protocol) server that searches arXiv, stores research papers by topic, and exposes tools and resources for structured academic discovery and analysis.

📋 Features

• 🔍 Search arXiv Papers: Search for academic papers on arXiv based on topics
• 📚 Store Paper Information: Automatically organize and store paper metadata by topic
• 🔧 MCP Tools: Expose search and extraction tools via MCP protocol
• 📊 MCP Resources: Access stored papers through structured resources
• 💡 Prompt Templates: Pre-built prompts for research assistance

Available Tools

1. search_papers: Search for papers on arXiv and store their information
2. extract_info: Retrieve information about a specific paper by ID

Available Resources

1. papers://folders: List all available topic folders
2. papers://{topic}: Get detailed information about papers on a specific topic

🚀 Installation

Prerequisites

• Python 3.12 or higher
• pip or uv package manager

Option 1: Using pip (Recommended for Production)

# Clone the repository
git clone https://github.com/Muhammadyousafrana/mcp-arxiv-research-server.git
cd mcp-arxiv-research-server

# Create a virtual environment
python -m venv venv

# Activate the virtual environment
# On Windows:
venv\Scripts\activate
# On macOS/Linux:
source venv/bin/activate

# Install dependencies
pip install -r requirements.txt

Option 2: Using uv (Faster Alternative)

# Clone the repository
git clone https://github.com/Muhammadyousafrana/mcp-arxiv-research-server.git
cd mcp-arxiv-research-server

# Install uv if you haven't already
pip install uv

# Create virtual environment and install dependencies
uv venv
uv pip install -e .

🔄 Converting Dependencies from uv to pip

If you're using uv and want to convert your project to use pip instead, follow these steps:

Step 1: Generate requirements.txt from pyproject.toml

The project uses pyproject.toml to define dependencies. To convert to a requirements.txt file:

# Install uv if not already installed
pip install uv

# Compile pyproject.toml to requirements.txt
uv pip compile pyproject.toml -o requirements.txt

This command will:

• Read the dependencies from pyproject.toml
• Resolve all dependencies and their versions
• Create a pinned requirements.txt file with all dependencies

Step 2: Understanding runtime.txt

The runtime.txt file specifies the Python version for deployment platforms (like Heroku, Render, etc.):

python-3.12.3

Important Notes about runtime.txt:

1. Purpose: Tells deployment platforms which Python version to use
2. Format: Must be in the format python-X.Y.Z (e.g., python-3.12.3)
3. Location: Place in the root directory of your project
4. Compatibility: Ensure the version matches your pyproject.toml requirement (>=3.12)

Step 3: Using pip with requirements.txt

Once you have requirements.txt, you can install dependencies using standard pip:

# Create a virtual environment
python -m venv venv

# Activate virtual environment
source venv/bin/activate  # On macOS/Linux
# or
venv\Scripts\activate  # On Windows

# Install from requirements.txt
pip install -r requirements.txt

Step 4: Keeping Dependencies in Sync

When you add new dependencies:

Option A: Update pyproject.toml first (Recommended)

# 1. Edit pyproject.toml and add your dependency
# 2. Regenerate requirements.txt
uv pip compile pyproject.toml -o requirements.txt

# 3. Install the new dependencies
pip install -r requirements.txt

Option B: Direct pip install

# Install new package
pip install package-name

# Update requirements.txt
pip freeze > requirements.txt

📦 Project Structure

mcp-arxiv-research-server/
├── research_server.py      # Main MCP server implementation
├── main.py                 # Entry point
├── pyproject.toml          # Project configuration and dependencies
├── requirements.txt        # Pip-compatible dependency list
├── runtime.txt             # Python version specification
├── uv.lock                 # uv lock file (if using uv)
├── .python-version         # Python version for pyenv
├── papers/                 # Directory for stored papers (created at runtime)
│   └── {topic}/
│       └── papers_info.json
├── .gitignore
├── LICENSE
└── README.md

🎯 Usage

Starting the Server

python research_server.py

The server will start on port 8001 using SSE (Server-Sent Events) transport.

Example: Searching for Papers

# The MCP server exposes the following tool
search_papers(topic="machine learning", max_results=5)

This will:

1. Search arXiv for papers on "machine learning"
2. Store metadata in papers/machine_learning/papers_info.json
3. Return a list of paper IDs

Example: Extracting Paper Information

# Extract information about a specific paper
extract_info(paper_id="2301.12345")

Example: Accessing Resources

Access stored papers through MCP resources:

• papers://folders - List all available topics
• papers://machine_learning - Get all papers on machine learning

🔧 Configuration

Dependencies

The project has two main dependencies defined in pyproject.toml:

[project]
name = "mcp-arxiv-research-server"
version = "0.1.0"
description = "Add your description here"
readme = "README.md"
requires-python = ">=3.12"
dependencies = [
    "arxiv>=2.4.0",
    "mcp>=1.26.0",
]

Python Version

• Minimum Required: Python 3.12
• Runtime Version: Python 3.12.3 (as specified in runtime.txt)
• Specified in:
  • pyproject.toml: requires-python = ">=3.12"
  • runtime.txt: python-3.12.3
  • .python-version: 3.12 (for pyenv users)

📝 Data Storage

Papers are stored in the following structure:

papers/
└── {topic_name}/
    └── papers_info.json

Each papers_info.json contains:

{
  "paper_id": {
    "title": "Paper Title",
    "authors": ["Author 1", "Author 2"],
    "summary": "Abstract text...",
    "pdf_url": "https://arxiv.org/pdf/...",
    "published": "2024-01-15"
  }
}

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🔗 Resources

• arXiv API Documentation
• Model Context Protocol (MCP)
• FastMCP Documentation

🐛 Troubleshooting

Common Issues

1. Import Error for mcp.server.fastmcp

   • Ensure you have mcp>=1.26.0 installed
   • Try reinstalling: pip install --upgrade mcp

2. Python Version Mismatch

   • Check your Python version: python --version
   • Must be Python 3.12 or higher
   • Update runtime.txt if deploying to a platform

3. Papers Directory Not Found

   • The papers/ directory is created automatically
   • Ensure you have write permissions in the project directory

4. Dependency Conflicts

   • Delete venv/ and reinstall:
     复制

     rm -rf venv
     python -m venv venv
     source venv/bin/activate
     pip install -r requirements.txt
     

📊 Development

Using uv for Development

# Install in editable mode
uv pip install -e .

# Add a new dependency
# Edit pyproject.toml, then:
uv pip compile pyproject.toml -o requirements.txt
uv pip install -e .

Using pip for Development

# Install in editable mode
pip install -e .

# Add a new dependency
pip install new-package
pip freeze > requirements.txt

🚀 Deployment

When deploying to platforms like Heroku, Render, or Railway:

1. Ensure runtime.txt specifies the correct Python version
2. Use requirements.txt for dependency installation
3. Set environment variables if needed
4. Ensure the papers/ directory is writable (or configure persistent storage)

Example Heroku Deployment

# Create Heroku app
heroku create your-app-name

# Ensure files are present
# - runtime.txt (python-3.12.3)
# - requirements.txt

# Deploy
git push heroku main

Made with ❤️ for the research community