Skip to main content
Glama
Sign In
rxiv

rxiv

by sakzgupzzz
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Remote

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault

No arguments

Instructions

Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.

This server publishes no instructions, or was last inspected before Glama recorded them.

Capabilities

Features and capabilities supported by this server

Protocol revision2025-11-25

CapabilityDetails
tools
{
  "listChanged": true
}

Tools

Functions exposed to the LLM to take actions

NameDescription
arxiv_searchA

Search arXiv papers. query can be plain text (words are AND-ed across title/abstract/authors) or raw arXiv syntax with field prefixes and booleans, e.g. ti:"protein language model" AND au:rives, abs:diffusion ANDNOT cat:cs.CV. Prefixes: ti (title), au (author), abs (abstract), cat (category), all. Combine with category (e.g. cs.LG, q-bio.BM, stat.ML) and a submitted-date window. Returns metadata and abstracts; use arxiv_get_full_text to read a paper.

arxiv_get_paperA

Fetch metadata and abstract for one or more arXiv papers by ID. Accepts 2410.12345, 2410.12345v2, arXiv:2410.12345, old-style hep-th/9901001, or any arxiv.org abs/pdf/html URL.

arxiv_get_full_textA

Return the full text of an arXiv paper as plain text (from arXiv's HTML rendering when available, otherwise extracted from the PDF). Long papers are paged: check truncated and pass next_offset as offset to continue.

biorxiv_searchA

Keyword search over bioRxiv and medRxiv preprints (via the Europe PMC index, which covers both servers). query can be plain text or Europe PMC syntax, e.g. TITLE:"single-cell" AND AUTH:"Regev A", (organoid OR spheroid) AND ABSTRACT:"drug screening". Filter by server and first-posted date; sort by relevance, date, or citation count. Returns abstracts, DOIs, and whether the published journal version exists. Use biorxiv_get_preprint for version history and biorxiv_get_full_text to read a paper.

biorxiv_get_preprintA

Fetch a preprint by DOI from the official bioRxiv/medRxiv API: title, authors, category, license, abstract, every posted version with dates, and the published journal version if one exists. Accepts a bare DOI (10.1101/2024.01.01.573711), a doi.org link, or a biorxiv.org/medrxiv.org content URL (version suffix like v2 is honoured). Auto-detects the server unless server is given.

biorxiv_list_recentA

List preprints posted to bioRxiv or medRxiv in a date window, optionally filtered by subject category (e.g. neuroscience, bioinformatics, genomics, cell biology, immunology; medRxiv: epidemiology, oncology, infectious diseases). Defaults to today. Pages hold 30 records; pass next_cursor as cursor to continue. Use this for browsing what is new; use biorxiv_search for keyword queries.

biorxiv_get_full_textA

Return the full text of a bioRxiv or medRxiv preprint as plain text, using the copy held by Europe PMC (JATS XML, falling back to PDF). Accepts a DOI, a preprint URL, or a Europe PMC PPR id. Full text usually appears a few days after posting; if it is not available yet the error says so. Long papers are paged: check truncated and pass next_offset as offset to continue.

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

Deploy Server
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

A4.1/5.0

Scored across 7 tools

Disambiguation4/5

The arxiv_/biorxiv_ prefixes cleanly partition the tool set into two families, and within each family search / metadata / full-text are clearly different actions. The only mild overlaps are biorxiv_search vs biorxiv_list_recent (browsing vs keyword) and the two metadata fetchers, but the descriptions explicitly disambiguate them.

Naming Consistency4/5

Names consistently follow prefix_verb_noun (arxiv_get_paper, biorxiv_get_full_text, biorxiv_list_recent), so the pattern is predictable. The one asymmetry is the metadata-fetch verb: get_paper for arXiv vs get_preprint for bioRxiv, which is a minor, defensible deviation given the different resource types.

Tool Count5/5

Seven tools is well-scoped for a two-repository preprint server: three arXiv operations and four bioRxiv/medRxiv operations, each earning its place with no redundancy.

Completeness4/5

Search, metadata retrieval, full-text retrieval, and pagination are covered for both sources, plus a recent-listings feed for bioRxiv/medRxiv. The main gap is the absence of an equivalent 'list recent' or date-browse tool for arXiv (and no citation/reference traversal), though agents can approximate arXiv browsing with search date filters.

Maintenance

ActivityMaintained
ResponsivenessNo issues

Resources