About MyArxivDB_MCP MCP server
MyArxivDB_MCP is an MCP server for researchers organizing arXiv papers in a personal PostgreSQL database. It crawls papers, supports semantic search and project assignment, and generates literature-review sections with LLM prompts.
Key features
- 01Crawl arXiv metadata and PDFs
- 02Embed abstracts with Pinecone
- 03Store papers and projects in PostgreSQL with pgvector
- 04Semantic search for project literature
- 05Generate Related Work sections
Use cases
- 01Collect arXiv papers by ID or URL
- 02Find papers relevant to a research project
- 03Draft a Related Work section
MCP support
Provides paper crawling, storage, semantic retrieval, and literature-review generation capabilities to MCP clients. The README describes it primarily for Claude Desktop, with possible use by other MCP clients.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project README🧠 Organizing Research Papers with MCP Server
This repository provides a MCP server to organize personal research papers using an MCP (Modular Command Platform) server and semantic search. The platform enables efficient paper retrieval, automatic project assignment, and assistance in writing literature review sections using LLMs.
This server is mainly designed for Claude Desktop but may also work well with other MCP clients.
This project was carried out as part of the term project for the BKMS1 course(@SNU GSDS).
⚡️ Quickstart
The following quickstart guide is based on an Apple Silicon MacBook.
- Install uv
curl -LsSf https://astral.sh/uv/install.sh | sh
- clone the repository
git clone https://github.com/Jongbin-kr/MyArxivDB_MCP.git
- Intsall dependencies & activate the virtual environment
uv sync
source /.venv/bin/activate
- set up environment variables at
.envfile.
# .env
PINECONE_API_KEY = "YOUR PINECONE_API_KEY"
DB_NAME = "YOUR_DB_NAME"
DB_USER = "YOUR_DB_USERNAME"
DB_PASSWORD = "YOUR_DB_PASSWORD"
DB_HOST = "localhost"
DB_PORT = 4444
-
Intsall MCP server at Claude Desktop
mcp install server.py -
Done! The Claude desktop app will automatically detect the MCP server and you can start using i!
📌 Motivation
Researchers frequently accumulate large numbers of papers but lack tools to systematically organize them by topic or project. BKMS aims to:
- Automatically assign new papers to relevant projects using embeddings
- Allow semantic search for project-specific literature
- Assist in drafting sections like “Related Work” using LLMs
🛠️ Main functions
Our MCP server supports the following core capabilities:
- Crawling metadata and PDFs from arXiv using ID or URL using ArXiv API
- Embedding abstracts using Pinecone API(
llama-text-embed-v2) - Storing papers and projects in a PostgreSQL + pgvector DB
- Generating "Related Work" sections via LLM prompts
🎥 Workflow & Demo video
You can see our PPT and demo video in assets folder.
Brief overview of our project workflow and DB schema is as follows.
👨👩👧👦 Team
- 박연진
- 원종빈
- 정예준
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.