Scopus Plus MCP

What Scopus Plus MCP does

Scopus Plus MCP MCP server exposes literature-search and bibliometric workflows through the Model Context Protocol. It can query Scopus, use OpenAlex as an alternative source for supported analyses, and call Crossref, Semantic Scholar, and open-access sources when particular operations require them. The project is listed in the 🔬 Research category and is intended for MCP clients that can pass natural-language requests to its tools.

The server covers record discovery, author lookup, abstracts, identifiers, references, citing papers, and Elsevier full-text search. Full-text search can find papers using a term in the body and report where and how often the term appears. Bibliometric functions include publication counts, topic landscapes and profiles, thematic change over time, journal metrics, journal discovery, and BibTeX generation.

How it works

An MCP client interprets a request and selects one or more of the server's tools. Scopus calls use the user's Elsevier API key. OpenAlex can be selected with source="openalex" and does not require a key. Results are returned as a conversational summary, while complete datasets and analysis artifacts can be saved for use with Pajek, VOSviewer, Gephi, or Zotero.

Citation-analysis tools support bibliographic coupling, co-citation, citation lineages, citation networks, historiographs, RPYS, and research-front analysis. Audit tools can resolve citing sets, retrieve citation contexts, compare index coverage, check retractions, and examine path transmission or coding agreement. Cached responses and background jobs support longer-running operations; job status and job result tools retrieve asynchronous work.

Setup and configuration

Scopus Plus MCP MCP server requires an API key from the Elsevier Developer Portal for Scopus access. OpenAlex-based operations do not require a key. Claude Desktop users can install the .mcpb package from a release and enter the key during installation. Claude Code users can install the published plugin and expose the key through SCOPUS_API_KEY or an operating-system secret store.

Other MCP clients can run the server with uvx; the README identifies uv as the prerequisite. The project requires Python 3.10 or newer. The material provided does not specify additional required environment variables.

Tools and capabilities

The documented tool families are:

• Search and records: Scopus and cross-source search, full-text search, abstracts, identifiers, authors, profiles, and full text.
• Citations: references and citing papers.
• Networks: coupling, co-citation, lineages, networks, historiographs, RPYS, and research fronts.
• Audit: citer resolution, citation context, path transmission, coding agreement, coverage, and retraction checks.
• Bibliometrics: counts, topic analysis, thematic evolution, journal metrics, journal discovery, BibTeX, and record import.
• Diagnostics and jobs: connection checks, quota status, server information, job status, and job results.

Scopus Plus MCP MCP server can also export analysis data in formats including JSON, CSV, GraphML, Pajek, PNG, HTML, and BibTeX.

Limitations and notes

Scopus features depend on the user's Elsevier access and supported subscription permissions. The server includes a diagnostic operation that reports which tools the available access supports and explains failed access. OpenAlex provides an alternative for supported analyses, but the provided material does not state that every Scopus operation has an OpenAlex equivalent. Long jobs may run asynchronously, and full outputs are written to files rather than appearing entirely in the chat response.