@cyanheads/biorxiv-mcp-server

Public Hosted Server: https://biorxiv.caseyjhand.com/mcp

Overview

bioRxiv and medRxiv preprint metadata and full text, searchable via EuropePMC. Fetch preprints by DOI, browse by date interval or subject category, search by keyword and author, resolve journal-publication crosswalks, and extract full text from the rendered article page, from any MCP client. Runs as a stdio process, a local Streamable HTTP server, or the public hosted endpoint above.

Tools

Related MCP server: medRxiv-MCP-Server

Capability reference

biorxiv_get_preprint tool

• Takes up to 10 DOIs per call, bare or pasted (https://doi.org/…, doi:…, a biorxiv.org/medrxiv.org article URL, a vN or .full suffix), scoped to biorxiv, medrxiv, or both (default both)

• Each preprint returns its full revision history in revisions[] — title, authors, abstract, category, license, awards, jatsxmlUrl, and publishedJournalDoi once accepted. Funder names are left out because api.biorxiv.org attributes them to unrelated organizations

• A DOI that fails lands in failed[] with a reason (not_found, invalid_doi_format, upstream_unavailable, rate_limited) and a retryable flag; not_found only when every attempted server answered

biorxiv_list_recent tool

• Takes a start_date/end_date interval, server (default both), and optional category (a value from biorxiv_list_categories) and bioRxiv-only funder (a ROR ID) filters; a filter the API ignored or cannot apply raises invalid_category or invalid_funder rather than returning an unfiltered or empty page

• Pages of 30 advance with an integer cursor; pagination is per server ({ biorxiv: { cursor, total }, medrxiv: { cursor, total } }), a cursor past the last page is marked exhausted: true, and a server that did not answer is named in failed[]

• Abstracts are omitted by default (about three quarters of a page); include_abstract: true adds them

biorxiv_search_preprints tool

• Takes query and/or author, optional date_from/date_to and server (default both); up to 100 results per page (default 25), paged with cursor_mark

• EuropePMC ranks the matches (it indexes new preprints within 1–2 days of posting) and the bioRxiv/medRxiv API enriches them with the same fields as biorxiv_get_preprint; a record left with EuropePMC metadata only carries enriched: false and an enrichment_error (service_error, rate_limited, not_found), with partial_results set

• Abstracts are included by default; include_abstract: false drops them for a response about a third the size

biorxiv_get_published_version tool

• Takes one preprint DOI and server (default both — the two servers share their DOI prefixes); 10.64898/ DOIs, which the /pubs crosswalk cannot key on, resolve through the preprint's own journal DOI

• Returns journal DOI, journal name, published date, corresponding-author institution, and the server that answered; no server answering raises a retryable upstream_unavailable or rate_limited, never doi_not_found

biorxiv_get_fulltext tool

• Takes a DOI, server (default both), and an optional version (or a vN suffix — the two must agree; a version the preprint lacks raises version_not_found); extracts Markdown from the rendered HTML article page, since there is no keyless JATS source, and names the answering server

• Pages via offset/limit characters (default 20,000, max 50,000), reporting totalChars, remainingChars, hasMore, and wordCount; the extracted article is cached per version, so paging costs one origin fetch

• PDF-only preprints and blocked pages raise fulltext_unavailable, routing to biorxiv_get_preprint

biorxiv_list_categories tool

• No API call — a static list of 25 bioRxiv and 51 medRxiv categories, limited to the ones the listing API filters on

• Use it to pick a category for biorxiv_list_recent

Features

Built on @cyanheads/mcp-ts-core: stdio and Streamable HTTP transports, pluggable auth (none / jwt / oauth), swappable storage (in-memory, filesystem, Supabase, Cloudflare KV/R2/D1), structured logging with optional OpenTelemetry tracing.

bioRxiv-specific:

• BiorxivApiService wraps api.biorxiv.org — details, publications, and crosswalk endpoints with retry and exponential backoff; a 429 is classified as a retryable rate_limited error carrying the parsed Retry-After wait, with the upstream response body kept out of the payload

• EuropePmcService wraps the EuropePMC search endpoint for relevance-ranked keyword and/or author results, classified the same way on a 429

• BiorxivFullTextService fetches and extracts Markdown from the rendered HTML article pages on www.biorxiv.org / www.medrxiv.org — a distinct origin from the JSON API

• Two-server fan-out via Promise.allSettled — both biorxiv and medrxiv queried in parallel when server="both", results merged and deduplicated by DOI

• Polite User-Agent header including a mailto address (BIORXIV_MAILTO env var) per Cold Spring Harbor Lab API guidelines

• Pairs with pubmed-mcp-server (post-publication), openalex-mcp-server (citation analytics), and crossref-mcp-server (DOI metadata)

Agent-friendly output:

• Graceful partial failure — per-DOI and per-server failures land in failed[] with a typed reason and retryable flag instead of aborting the whole batch or listing call

• Rate-limit transparency — a 429 from any upstream surfaces as reason: "rate_limited" carrying the origin's parsed retryAfter wait, distinguished from a generic upstream_unavailable

• Discriminated enrichment outputs — biorxiv_search_preprints results carry enriched plus a typed enrichment_error (service_error / rate_limited / not_found) so callers branch on data, not string parsing

• Paging and pagination state — exhausted cursors are flagged as an out-of-range artifact rather than an empty interval, and biorxiv_get_fulltext reports totalChars / remainingChars / hasMore for chunked reads

• Clean titles and abstracts — Highwire export markup is resolved to plain text on both surfaces: symbol placeholders ({beta}, {+/-}, [≥]) become their characters, structured-abstract headings read Results: …, list items read • …, and figure and table blocks are dropped. The Markdown in content[] escapes upstream text so it renders as written

Getting started

Public Hosted Instance

A public instance is available at https://biorxiv.caseyjhand.com/mcp — no installation required. Point any MCP client at it via Streamable HTTP:

{
  "mcpServers": {
    "biorxiv-mcp-server": {
      "type": "streamable-http",
      "url": "https://biorxiv.caseyjhand.com/mcp"
    }
  }
}

Self-Hosted / Local

Add the following to your MCP client configuration file.

{
  "mcpServers": {
    "biorxiv-mcp-server": {
      "type": "stdio",
      "command": "bunx",
      "args": ["@cyanheads/biorxiv-mcp-server@latest"],
      "env": {
        "MCP_TRANSPORT_TYPE": "stdio",
        "MCP_LOG_LEVEL": "info",
        "BIORXIV_MAILTO": "your@email.com"
      }
    }
  }
}

Or with npx (no Bun required):

{
  "mcpServers": {
    "biorxiv-mcp-server": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "@cyanheads/biorxiv-mcp-server@latest"],
      "env": {
        "MCP_TRANSPORT_TYPE": "stdio",
        "MCP_LOG_LEVEL": "info",
        "BIORXIV_MAILTO": "your@email.com"
      }
    }
  }
}

Or with Docker:

{
  "mcpServers": {
    "biorxiv-mcp-server": {
      "type": "stdio",
      "command": "docker",
      "args": ["run", "-i", "--rm", "-e", "MCP_TRANSPORT_TYPE=stdio", "-e", "BIORXIV_MAILTO=your@email.com", "ghcr.io/cyanheads/biorxiv-mcp-server:latest"]
    }
  }
}

For Streamable HTTP, set the transport and start the server:

MCP_TRANSPORT_TYPE=http MCP_HTTP_PORT=3010 BIORXIV_MAILTO=your@email.com bun run start:http
# Server listens at http://localhost:3010/mcp

Prerequisites

• Bun v1.4.0 or higher (or Node.js v24+).

Installation

1. Clone the repository:

git clone https://github.com/cyanheads/biorxiv-mcp-server.git

2. Navigate into the directory:

cd biorxiv-mcp-server

3. Install dependencies:

bun install

4. Configure environment:

cp .env.example .env
# optionally set BIORXIV_MAILTO for polite API access

Configuration

All configuration is validated at startup via Zod schemas in src/config/server-config.ts.

See .env.example for the full list of optional overrides.

Running the server

Local development

• Build and run:

  # One-time build
  bun run rebuild
  
  # Run the built server
  bun run start:stdio
  # or
  bun run start:http

• Run checks and tests:

  bun run devcheck   # Lint, format, typecheck, security
  bun run test       # Vitest test suite
  bun run lint:mcp   # Validate MCP definitions against spec

Docker

docker build -t biorxiv-mcp-server .
docker run --rm -e BIORXIV_MAILTO=your@email.com -p 3010:3010 biorxiv-mcp-server

The Dockerfile defaults to HTTP transport, stateless session mode, and logs to /var/log/biorxiv-mcp-server. OpenTelemetry peer dependencies are installed by default — build with --build-arg OTEL_ENABLED=false to omit them.

Project structure

Development guide

See CLAUDE.md for development guidelines and architectural rules. The short version:

• Handlers throw, framework catches — no try/catch in tool logic

• Use ctx.log for request-scoped logging, ctx.state for tenant-scoped storage

• Register new tools via the barrel in src/mcp-server/tools/definitions/index.ts

• Wrap external API calls: validate raw → normalize to domain type → return output schema; never fabricate missing fields

Contributing

Issues are welcome. Run checks and tests before submitting:

bun run devcheck
bun run test

License

Apache-2.0 — see LICENSE for details.