Research MCP Server

A powerful Model Context Protocol (MCP) server that provides academic research capabilities through arXiv integration. This server enables AI assistants to search, retrieve, and analyze academic papers seamlessly.

Features

• Paper Search - Search arXiv by topic, author, or keywords with customizable result limits
• Paper Analysis - Extract detailed information from specific research papers
• Resource Access - Browse available topics and access organized paper collections
• Smart Prompts - Generate structured research prompts for comprehensive analysis
• MCP Compatible - Full Model Context Protocol implementation for seamless AI integration
• Fast & Reliable - Efficient arXiv API integration with robust error handling
• Easy Integration - Simple setup for use with MCP clients and AI assistants

Related Projects

• research-mcp-client - A Claude AI chatbot that uses this server for intelligent research assistance

Prerequisites

• Python 3.8+
• arxiv package (>=2.1.0) - Official Python wrapper for arXiv.org
• Internet connection - Required for arxiv package to access arXiv.org
• MCP client (like the research-mcp-client)

Quick Start

1. Clone the Repository

git clone https://github.com/spalit2025/research-mcp-server.git
cd research-mcp-server

2. Set Up Virtual Environment

python3 -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install Dependencies

pip install -r requirements.txt

4. Run the Server

python research_server.py

Usage Examples

As MCP Server

The server runs as a stdio-based MCP server, perfect for integration with AI assistants:

python research_server.py

Available Tools

search_papers

Search for academic papers on arXiv by topic.

Parameters:

• topic (string): Search query for papers
• max_results (integer, optional): Maximum number of results (default: 5)

Example:

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

extract_info

Extract detailed information from a specific paper.

Parameters:

• paper_id (string): The paper identifier from search results

Example:

{
  "paper_id": "paper_1"
}

Available Resources

papers://folders

Lists all available topic folders with papers stored locally.

Returns: Markdown-formatted list of available research topics

Usage: Access this resource to see what topics have been previously searched and have papers available.

papers://{topic}

Get detailed information about all papers for a specific topic.

Parameters:

• topic (string): The research topic to retrieve papers for

Returns: Comprehensive markdown document with paper details including titles, authors, summaries, and PDF links

Example: papers://machine_learning or papers://quantum_computing

Available Prompts

generate_search_prompt

Generate a structured prompt for comprehensive academic research on a specific topic.

Parameters:

• topic (string): The research topic to generate a prompt for
• num_papers (integer, optional): Number of papers to search for (default: 5)

Returns: Detailed prompt that guides systematic research including:

• Paper search instructions
• Information extraction guidelines
• Analysis and synthesis requirements
• Structured output formatting

Example:

{
  "topic": "neural networks",
  "num_papers": 8
}

Development & Testing

Test with MCP Inspector

The MCP Inspector provides a web interface for testing your server:

npx @modelcontextprotocol/inspector python research_server.py

Then open http://localhost:6274 in your browser to interact with the server.

Standalone Testing

Test the server functions directly:

python -c "
import research_server
result = research_server.search_papers('artificial intelligence', 3)
print(result)
"

Development Setup

The main development tool you'll need is the MCP Inspector for testing:

# Install MCP Inspector globally
npm install -g @modelcontextprotocol/inspector

# Or use npx to run it directly
npx @modelcontextprotocol/inspector python research_server.py

Project Structure

research-mcp-server/
├── research_server.py      # Main MCP server implementation
├── requirements.txt        # Production dependencies
├── requirements-dev.txt    # Development dependencies
├── .gitignore             # Git ignore rules
├── LICENSE                # MIT License
└── README.md              # This file

How It Works

1. MCP Server - Implements the Model Context Protocol for tool communication
2. arXiv Integration - Uses the arxiv Python package to search and retrieve papers from arXiv.org
3. Tool Registration - Exposes search_papers and extract_info as MCP tools
4. Resource Management - Provides organized access to paper collections via resources
5. Prompt Generation - Creates structured research prompts for systematic analysis
6. Data Processing - Formats and structures paper information for AI consumption
7. Error Handling - Robust error management for network and API issues

Configuration

Customizing Search Parameters

You can modify the default search behavior in research_server.py:

# Adjust default result limits
DEFAULT_MAX_RESULTS = 5

# Modify search categories or sorting
# See arXiv API documentation for options

Integration with Clients

This server is designed to work with MCP clients. Configure your client to connect using:

server_params = StdioServerParameters(
    command="python",
    args=["/path/to/research-mcp-server/research_server.py"],
    env=None,
)

Troubleshooting

Connection Issues

• ✅ Ensure Python 3.8+ is installed and accessible
• ✅ Verify all dependencies are installed correctly
• ✅ Check that the server starts without errors

arXiv Package Issues

• ✅ Verify arxiv package installation: pip show arxiv
• ✅ Test package import: python -c "import arxiv; print('arxiv package version:', arxiv.__version__)"
• ✅ Check arXiv.org accessibility in your browser
• ✅ Ensure no firewall blocking connections to arXiv.org

Search Problems

• ✅ Try simpler search terms if no results found
• ✅ Check arXiv.org directly to verify paper availability
• ✅ Reduce max_results if getting timeout errors

Common Solutions

# Reinstall dependencies
pip install -r requirements.txt --force-reinstall

# Test arXiv connectivity
python -c "import arxiv; print('arXiv accessible')"

# Check server startup
python research_server.py --help

Development

Adding New Features

1. Implement new tool functions in research_server.py
2. Register tools with the MCP server
3. Update tool schemas and documentation
4. Test with MCP Inspector

Code Structure

• Tool Functions - Implement the actual research functionality (search_papers, extract_info)
• Resource Functions - Provide organized access to stored paper data (get_available_folders, get_topic_papers)
• Prompt Functions - Generate structured research prompts (generate_search_prompt)
• MCP Integration - Handle protocol communication and tool registration
• Error Handling - Manage API errors and edge cases
• Data Formatting - Structure responses for optimal AI consumption

Dependencies

Production

• arxiv >=2.1.0 - Official Python package for arXiv.org integration
• mcp >=1.0.0 - Model Context Protocol server implementation

Development & Testing

• MCP Inspector - Web-based tool for testing MCP servers

  npx @modelcontextprotocol/inspector python research_server.py
  

Note: Additional development tools like pytest, black, and flake8 are listed in requirements-dev.txt but not currently configured for this project.

License

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

Contributing

1. Fork the repository
2. Create a feature branch (git checkout -b feature/amazing-feature)
3. Make your changes and add tests
4. Ensure code follows style guidelines
5. Commit your changes (git commit -m 'Add some amazing feature')
6. Push to the branch (git push origin feature/amazing-feature)
7. Open a Pull Request

Development Guidelines

• Follow PEP 8 style guidelines
• Test your changes with MCP Inspector
• Update documentation for any changes
• Ensure MCP compatibility is maintained