ArXiv Research Assistant MCP Server

Installation

Using Smithery (Recommended)

The easiest way to install ArXiv Research Assistant is via Smithery:

npx -y @smithery/cli install arxiv-mcp-server --client claude

Manual Installation

You can install using uv:

uv tool install arxiv-mcp-server

For development purposes:

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

# Create and activate virtual environment
uv venv
source .venv/bin/activate

# Install with test dependencies
uv pip install -e ".[test]"

Configuration

MCP Client Configuration

Add the following configuration to your MCP client config file:

{
    "mcpServers": {
        "arxiv-mcp-server": {
            "command": "uv",
            "args": [
                "tool",
                "run",
                "arxiv-mcp-server",
                "--storage-path", "/path/to/paper/storage"
            ]
        }
    }
}

For development:

{
    "mcpServers": {
        "arxiv-mcp-server": {
            "command": "uv",
            "args": [
                "--directory",
                "path/to/cloned/arxiv-mcp-server",
                "run",
                "arxiv-mcp-server",
                "--storage-path", "/path/to/paper/storage"
            ]
        }
    }
}

Environment Variables

You can configure the server using the following environment variables:

| Variable | Purpose | Default | | --- | --- | --- | | ARXIV_STORAGE_PATH | Paper storage location | ~/.arxiv-mcp-server/papers |

Usage

Once installed and configured, you can use the ArXiv Research Assistant through your AI assistant. The server provides tools for searching papers, downloading them, listing your downloaded papers, and reading their content.

Example Workflows

1. Finding recent papers on a topic:

   • Search for papers with specific keywords, date range, and categories
   • Review the search results to identify relevant papers
   • Download papers of interest
   • Read and analyze the papers

2. Comprehensive paper analysis:

   • Use the deep-paper-analysis prompt with a paper ID
   • Get a structured analysis including executive summary, methodology review, and implications

3. Literature review:

   • Search for papers across multiple related topics
   • Download and organize papers by theme
   • Use the reading tools to extract key information from each paper

The server maintains a local cache of downloaded papers, so subsequent requests for the same papers will be faster.

Testing

To run the test suite:

python -m pytest

This will verify that all components of the server are functioning correctly.