The bioRxiv MCP Server acts as a bridge between AI assistants and bioRxiv's preprint repository using the Model Context Protocol (MCP). It enables AI models to search for biology preprints and access their metadata in a programmatic way. Core capabilities include keyword and advanced search, rapid retrieval of paper metadata, and access to full paper content for offline reading. Papers can be downloaded and stored locally to accelerate subsequent queries, and a simple MCP interface exposes tools for searching, retrieving metadata, and listing downloaded papers. The server is implemented in Python and leverages FastMCP, with prerequisites including Python 3.10+ and asyncio support. The project provides integration options with multiple clients (Claude Desktop, Cursor, Windsurf, CLine) via Smithery, and includes example commands and configuration snippets for seamless setup. Project structure features biorxiv_server.py as the main MCP server and biorxiv_web_search.py for the web scraping logic. This tool aims to support scientific research by offering structured access to bioRxiv metadata and content, enabling researchers to query, filter, and analyze preprints through programmable prompts and workflows.