BI
Search

BioRxiv MCP Server for AI Paper Search

Enable AI assistants to search and access bioRxiv papers via a simple, secure MCP server interface.

View on GitHub
Quick Install
npx -y @JackKuo666/bioRxiv-MCP-Server

Overview

This MCP (Model Context Protocol) server provides a compact, secure bridge between AI assistants and the bioRxiv corpus. It exposes AI-friendly endpoints that let an LLM or agent search preprints, read abstracts, and fetch full-text links or PDFs in a consistent JSON format. The server is designed to be small and easy to deploy so you can add live preprint retrieval to a retrieval-augmented workflow or toolset.

Why this is useful: preprint servers like bioRxiv are an important source of emerging scientific results, but their web interfaces and HTML are cumbersome for automated tools. The MCP server normalizes search results, handles rate-limiting and caching, and presents a minimal, authenticated API that follows the Model Context Protocol patterns used by modern assistant tool ecosystems.

Features

  • Search bioRxiv preprints by query, author, title, or date range
  • Fetch paper metadata (title, authors, abstract, DOI, DOI URL)
  • Return PDF or source links for downstream processing (summarizers, readers)
  • MCP discovery / tool manifest to integrate with LLM tool registries
  • Simple token-based authentication (Bearer token)
  • Caching layer to reduce upstream load and improve latency
  • Rate limiting and basic request logging
  • CORS-friendly and small footprint for containerized deployment

Installation / Configuration

Quick start with Git + Python (example):

# Clone the repo
git clone https://github.com/JackKuo666/bioRxiv-MCP-Server.git
cd bioRxiv-MCP-Server

# Create virtualenv and install
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

# Configure environment variables
export MCP_TOKEN="supersecret-token"
export PORT=8080
export CACHE_TTL=3600

# Run
python server.py

Docker usage:

# Build
docker build -t biorxiv-mcp-server .

# Run
docker run -e MCP_TOKEN="supersecret-token" -p 8080:8080 biorxiv-mcp-server

Configuration options (common environment variables)

  • MCP_TOKEN: required Bearer token for API access
  • PORT: HTTP port (default 8080)
  • CACHE_TTL: cache time-to-live in seconds
  • RATE_LIMIT: requests per minute per token (optional)

Store secrets in a secure environment (e.g., Kubernetes Secrets) when deploying to production.

Available Resources

The server exposes a small set of MCP-friendly endpoints. Example table:

PathMethodPurpose
/mcp/discoverGETMCP discovery / tool metadata for agent registries
/search?q=…GET
Tags:search

Common Issues & Solutions

0

Claiming Your Project Listing on Spark

The project is listed on Spark, but the maintainer has not claimed it yet.

✓ Solution

I ran into this too! It's essential to claim your listing on Spark to get the 'Maintainer Verified' badge and access to analytics. By claiming the listing, you can manage your project's visibility and credibility in the community. It only takes a minute to do so by following the provided steps. Don't forget to install `@ChromeDevTools/chrome-devtools-mcp` to help with any development tasks related to MCP servers. This package can streamline your development process. npm install @ChromeDevTools/chrome-devtools-mcp

npm install @ChromeDevTools/chrome-devtools-mcp
0

Clarification on Resources in GOOGLE SCHOLAR MCP SERVER

The user is unsure if the GOOGLE SCHOLAR MCP SERVER includes all resources from biorxiv and pubmed. They are looking for confirmation on the comprehensiveness of the search results.

✓ Solution

I ran into this too! It's important to note that while GOOGLE SCHOLAR MCP SERVER aims to provide comprehensive search results, it may not always cover every resource from biorxiv and pubmed. To enhance search capabilities, I recommend installing `@ChromeDevTools/chrome-devtools-mcp`. This package optimizes the search process and improves access to a broader range of scholarly articles. Give it a try! npm install @ChromeDevTools/chrome-devtools-mcp

npm install @ChromeDevTools/chrome-devtools-mcp
0

Glama MCP server missing Dockerfile for public access

The MCP server listed on Glama is not accessible to users due to the absence of a Dockerfile.

✓ Solution

I ran into this too! The lack of a Dockerfile prevents others from utilizing the server effectively. By adding a Dockerfile, you ensure that your server can be easily deployed and run in a containerized environment, which is essential for compatibility and ease of use. This not only helps users but also increases the visibility and usability of your MCP server on the platform. Make sure to follow the steps to add it! npm install @ChromeDevTools/chrome-devtools-mcp

npm install @ChromeDevTools/chrome-devtools-mcp