FreeHow credits work →

Research data MCP for Cursor

Add one entry to ~/.cursor/mcp.json and sign in. Cursor can then bring datasets, gene and protein records and occurrence data into the analysis project you have open.

Connect your agentSee all capabilities
ClaudeCodexChatGPTCursorGemini CLIAntigravityWorkBuddyTraeCodeBuddyCherry StudioQoderQwen CodeKimi Code

How to connect ScholarMuse to Cursor

Use the install link or edit one file; do one or the other. You can also paste “Read https://scholarmuse.ai/install/agent.md and follow the instructions to connect ScholarMuse” into Cursor’s agent and let it edit the file.

Full connect guide →

  1. Step 1

    Option A: open the install link

    Paste this link into your browser’s address bar. Cursor opens and asks whether to install the server.

    cursor://anysphere.cursor-deeplink/mcp/install?name=scholarmuse&config=eyJ1cmwiOiJodHRwczovL3NjaG9sYXJtdXNlLmFpL21jcCJ9

  2. Step 2

    Option B: edit mcp.json

    Put this in ~/.cursor/mcp.json in your home folder; create the file if it does not exist. If it already lists other servers, add only the scholarmuse entry under mcpServers.

    {
  "mcpServers": {
    "scholarmuse": {
      "url": "https://scholarmuse.ai/mcp"
    }
  }
}

  3. Step 3

    Start the sign-in

    Open Customize in Cursor’s sidebar, find scholarmuse in the list of MCP servers and make sure it is switched on. Cursor handles OAuth for servers that need it; start the sign-in from the server’s entry when Cursor offers it.

  4. Step 4

    Approve in your browser

    A scholarmuse.ai window opens. Sign in, approve access and choose which discipline packs Cursor may use. There is no API key to create or paste.

  5. Step 5

    Check that it works

    Ask the agent to list the ScholarMuse capabilities. By default Cursor asks for your approval before it runs an MCP tool.

What Cursor can bring into your project

Cursor works inside a project folder, so the list starts with data and records that end up as files next to your code. Life science sources come first; the last two entries are for the paper that the project leads to.

Find datasets

Find datasets

All disciplines · DataCite, Zenodo, Dryad +2

Gene & protein lookupNew

Gene & protein lookup

Life sciences & medicine · NCBI Gene, Ensembl, UniProt +5

Protein structures

Protein structures

Life sciences & medicine · RCSB PDB, AlphaFold DB

Species occurrence

Species occurrence

Earth & environment · GBIF

Economic & social statistics

Economic & social statistics

All disciplines · FRED, World Bank, IMF +5

Search papers

Search papers

All disciplines · OpenAlex, PubMed / Europe PMC, arXiv +5

Verify references

Verify references

All disciplines · Crossref, OpenAlex

Find datasets
Search DataCite, Zenodo, Dryad, Harvard Dataverse and Hugging Face from the editor, and keep the DOI and licence in the project notes.
Gene & protein lookup
Annotate a gene list in place: function, domains, pathways and tissue expression come back as fields Cursor can write into a table.
Protein structures
Get the experimental or predicted structure for a protein before Cursor writes the code that measures it.
Species occurrence
Openly licensed GBIF records with coordinates and dates, for a distribution map or a niche model.
Economic & social statistics
Covariates such as population or income, taken from the agency that publishes them and joined by date or region.
Search papers
Look up the paper that defines a method or a parameter value while the code is in front of you.
Verify references
Projects often hold a .bib file. Cursor can check its entries and correct the fields in the file.

Things to ask Cursor

Gene & protein lookup

You say to your agent

Use ScholarMuse to annotate the 40 genes in results/de_genes.csv with protein function and pathways, and add them as new columns. Keep the source of each annotation in a separate column.

Per-gene records with function, domains and pathways, which Cursor merges into the CSV.

Species occurrence

You say to your agent

Get GBIF occurrence records for Quercus robur in Spain since 2010, save them to data/occurrences.csv, and write an R script that maps them.

Occurrence records with coordinates, date and licence, written to disk for the mapping script.

Verify references

You say to your agent

Check every entry in paper/refs.bib with ScholarMuse. Fix wrong years, pages and DOIs in the file and list the entries you could not find.

Each entry comes back as verified, corrected or not found; Cursor shows the edits as a diff you can review.

Frequently asked questions

Cursor reads ~/.cursor/mcp.json for every project and .cursor/mcp.json for one project only. The global file suits most researchers. The entry holds only the server address, so it is also safe to keep in a project that you share.

ScholarMuse returns results to Cursor. Whether they become files is up to your request, so ask Cursor to save them, for example to data/, together with the source and licence that come with each result.

Yes. Cursor’s documentation says it asks for approval before using MCP tools by default, and you can open the arguments of a call before you accept it. Cursor’s own settings control which tools may run without asking.

It works, but the capabilities on this page pay off when there is a project to put the data in. For reading and writing alone, a chat assistant with ScholarMuse connected is simpler.

Every call your agent makes uses credits, and a call is charged only when it succeeds. There are no subscriptions. A new account starts with a one-time gift of 50 credits. The 6 web tools under /tools are free without an account, up to 5 runs a day.

Requests are used only to answer you. We do not train models on them.

Ready to make your agent a better scholar?

Connect once. Papers, data and figures from 50+ sources, in the agent you already use.

Works with compatible AI agents

ClaudeChatGPTCodexCursorWorkBuddyTraeGemini CLIAntigravityQoderCodeBuddyKimi CodeQwen CodeClineKiroWindsurfZedOpenCodeOpenClawDifyCherry StudioLM StudioGooseAmpContinueRoo CodeVS CodeJetBrainsWarpRaycastCopilotPerplexityAny MCP client

Data from OpenAlex · Crossref · PubMed · arXiv · NIH · UniProt · PDB · PubChem · NASA · World Bank · USGS · GBIF and more