rcuevass/MCP_server_example

MCP Research Server

A near production-ready Model Context Protocol (MCP) server for ArXiv paper research.

This repo is motivated by, and heavily dependent on the MCP: Build Rich-Context AI Apps with Anthropic short DeepLearning.AI course. It is an attempt to have a modularized version of the chatbot and with production-ready code.

Quick Start

1. Setup Environment:

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate
   pip install -e .
   cp .env.example .env
   

2. Run the Server:

   python -m mcp_research_server
   

3. Test with MCP Inspector:

   npx @modelcontextprotocol/inspector python -m mcp_research_server
   

Available Tools

• search_papers: Search ArXiv for papers on a topic
• extract_info: Get detailed information about a specific paper
• get_database_stats: View database statistics

Example Usage

# Search for papers
search_papers(topic="machine learning", max_results=10)

# Get paper details
extract_info(paper_id="2301.00001")

# View database statistics
get_database_stats()

Configuration

Edit .env file to customize:

• MCP_LOG_LEVEL: Logging level (DEBUG, INFO, WARNING, ERROR)
• MCP_ARXIV_MAX_RESULTS: Default number of search results
• MCP_SERVER_NAME: Server name for MCP

Integration with Claude Desktop

Add to Claude Desktop configuration:

{
  "mcpServers": {
    "research": {
      "command": "python",
      "args": ["-m", "mcp_research_server"],
      "cwd": "/path/to/mcp_research_server"
    }
  }
}

For more details, see the complete documentation in the project files.