Skip to main content
MCP server CX Codex

paper-search-mcp-openai-v2 — MCP server: search and fetch academic papers

The paper-search-mcp-openai-v2 MCP server for Claude Code searches arXiv, PubMed, bioRxiv, medRxiv, and other databases, and downloads PDFs to extract their text.

no license 7,042 installs updated 2025-10-31

Install into Claude Code · Platform CLI: remote server All methods and platforms ↓
claude mcp add --transport http paper-search-mcp-openai-v2 https://paper-search-mcp-openai-v2--titansneaker.run.tools
Open on GitHub

What paper-search-mcp-openai-v2 does

This MCP server adds paper search and retrieval tools to AI-powered code editors. Search across eight academic sources—arXiv for physics, math and CS; PubMed for biomedical literature; bioRxiv and medRxiv for preprints; Google Scholar for broad coverage; Semantic Scholar with year filters; IACR ePrint for cryptography; CrossRef for citations and DOIs. Each source has dedicated search, download, and text-extraction tools, plus a general search that queries multiple sources at once.

Installation varies by platform. Claude Code uses claude mcp add, Cursor requires JSON config, Codex and Gemini CLI accept command-line syntax. The server connects via HTTP to a remote endpoint at https://paper-search-mcp-openai-v2--titansneaker.run.tools. Internet access and connectivity to this host are required.

PDF download is available from arXiv, bioRxiv, medRxiv, IACR, and Semantic Scholar. PubMed and CrossRef provide metadata and DOIs only; access to full papers requires visiting the publisher's site directly.

What it helps with

  • gather papers on a topic from multiple repositories in one search
  • fetch and parse PDF text without leaving your editor
  • locate cryptographic or biomedical papers across specialized archives
  • retrieve citation metadata and DOIs for references

Example prompts for the agent

  • Search arXiv for recent papers on large language models and download the most relevant PDF
  • Find PubMed papers on immunotherapy published in the last three years
  • Fetch and read the full text of a bioRxiv preprint by its DOI

How to install paper-search-mcp-openai-v2

Platform CLI: remote server

claude mcp add --transport http paper-search-mcp-openai-v2 https://paper-search-mcp-openai-v2--titansneaker.run.tools

Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more

Questions about paper-search-mcp-openai-v2

How do I install paper-search-mcp-openai-v2 in Claude Code?

Run claude mcp add --transport http paper-search-mcp-openai-v2 https://paper-search-mcp-openai-v2--titansneaker.run.tools. Once added, search and download tools will be available to Claude.

Which academic sources does this server support?

Eight: arXiv, PubMed, bioRxiv, medRxiv, Google Scholar, Semantic Scholar, IACR ePrint, and CrossRef. Each has dedicated search and retrieval tools, plus a general search across multiple sources.

Can I download PDFs from all sources?

PDF download works with arXiv, bioRxiv, medRxiv, IACR, and Semantic Scholar. PubMed and CrossRef return metadata only; to access full papers, you must visit the publisher site.

Similar MCP servers

All MCP servers for AI and ML →

Project sponsors

Companies that keep this analytics open