MCP Research Server

Exposes arXiv-based research tools via MCP for discovering and extracting paper metadata.

python

0

GitHub Stars

python

Language

6 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "singhangad05-mcp_research": {
      "command": "python",
      "args": [
        "research_server.py"
      ]
    }
  }
}

You can run a simple MCP server that exposes research tools backed by arXiv. A lightweight client discovers these tools on the fly, decides when to call them, and feeds the results back for final reasoning. This setup helps you understand MCP end-to-end and experiment locally with a clear, hands-on workflow.

How to use

You’ll start the MCP client and server locally. The server exposes two tools that your client can call: one to search arXiv papers by topic and another to fetch stored metadata for a given paper. The client automatically discovers these tools, decides when to call them, and passes the results back to the reasoning model for the final answer.

How to install

Prerequisites: you need Python installed on your system. You’ll also run the client locally, which uses a local LLM runtime if you choose, but installation here focuses on the server setup and running both components.

python -m venv .venv
source .venv/bin/activate   # Windows: .venv\Scripts\activate
pip install -r requirements.txt

# Running MCP server
python research_server.py

# Running MCP client
python mcp_chatbot.py

# Optional: start an LLM runtime locally if your environment requires it (e.g., Ollama for the client runtime)
ollama serve  # ensure the local LLM server is running if you use it

Additional content

Notes about the workflow and common setup steps are outlined here to help you get started quickly and reliably.

Available tools

search_papers

Searches arXiv for papers by topic and stores metadata locally. Inputs: topic (string), max_results (int). Returns: List of arXiv paper IDs.

extract_info

Fetches stored metadata for a given paper_id. Inputs: paper_id (string). Returns: Title, authors, abstract, PDF link, publish date.