Semanticscholar

Official registry Other

Semantic Scholar Academic Graph MCP.

What can you do with Semanticscholar?

  • search_papers(query, year?, limit?) — papers matching a query: title,
  • get_paper(paper_id) — one paper by Semantic Scholar id, DOI:… or
  • get_paper_citations(paper_id, limit?) — the papers citing it.
  • get_author(name) — an author's papers and citation metrics.

Install

Add this to your client's MCP configuration (Claude Desktop: claude_desktop_config.json; Cursor: .cursor/mcp.json; VS Code and others use the same "mcpServers" shape).

Remote (streamable-http)
Claude Code
claude mcp add --transport http semanticscholar 'https://gateway.pipeworx.io/semanticscholar/mcp'
JSON configuration (Claude Desktop, Cursor, VS Code)
{
    "mcpServers": {
        "semanticscholar": {
            "type": "http",
            "url": "https://gateway.pipeworx.io/semanticscholar/mcp"
        }
    }
}
NPM package @pipeworx/mcp-semanticscholar
Claude Code
claude mcp add semanticscholar -e 'PIPEWORX_API_KEY=<your pipeworx_api_key>' -- npx -y @pipeworx/mcp-semanticscholar@0.2.0
JSON configuration (Claude Desktop, Cursor, VS Code)
{
    "mcpServers": {
        "semanticscholar": {
            "command": "npx",
            "args": [
                "-y",
                "@pipeworx/mcp-semanticscholar@0.2.0"
            ],
            "env": {
                "PIPEWORX_API_KEY": "<your pipeworx_api_key>"
            }
        }
    }
}

Environment variables

  • PIPEWORX_API_KEY secret Optional Pipeworx API key. Without it calls are anonymous and capped daily; a free key raises the cap.

Documentation

From the project's README on GitHub, refreshed weekly. Links open on GitHub.

Contents
  1. @pipeworx/semanticscholar
  2. Tools
  3. Auth
  4. Licence — zero-rated (fleet #1974)
  5. Data sources
  6. Quick Start
  7. No MCP client? Call it over HTTP
  8. Standalone (no gateway account)
  9. Using with ask_pipeworx
  10. More
  11. License

@pipeworx/semanticscholar

Semantic Scholar Academic Graph: search 200M+ papers, resolve a paper by ID, DOI or arXiv id, trace its citations, and look up authors with citation metrics and h-index.

Part of Pipeworx — an MCP gateway connecting AI agents to 1683+ live data sources.

Tools

  • search_papers(query, year?, limit?) — papers matching a query: title, year, authors, venue, citation count, DOI, open-access PDF, abstract.
  • get_paper(paper_id) — one paper by Semantic Scholar id, DOI:… or arXiv:….
  • get_paper_citations(paper_id, limit?) — the papers citing it.
  • get_author(name) — an author's papers and citation metrics.

Auth

Platform key (PLATFORM_SEMANTICSCHOLAR_KEY) sent as x-api-key. Without a key the public graph endpoints answer but rate-limit hard (429 under load).

Licence — zero-rated (fleet #1974)

The key was issued under the Semantic Scholar API License Agreement (https://api.semanticscholar.org/license/). §2(a) limits use to "legitimate, non-commercial, research and/or educational purposes" and requires that "any public use of Data must point back to Semantic Scholar at https://www.semanticscholar.org/ with a utm_source=api UTM parameter"; the Use Restrictions forbid "sell, lease, share, transfer, sublicense, commercialize … any Data obtained through the API" to third parties. Consequences:

  • zeroRated: true on the gateway entry: 0 credits on every tier, no volume-bracket slot.
  • Every response leads with attribution (the point-back), license (LicenseRef-SemanticScholar-API-License-Agreement, kind: vendor_terms, obligations attribution + non_commercial) and license_note, via attachLicense in @pipeworx/shared.
  • Zero-rating removes the fee. It does not license onward sharing: the caller's own use has to be non-commercial too, and the note says so.

Data sources

Quick Start

Add to your MCP client (Claude Desktop, Cursor, Windsurf, etc.):

{
  "mcpServers": {
    "semanticscholar": {
      "url": "https://gateway.pipeworx.io/semanticscholar/mcp"
    }
  }
}
What this endpoint actually serves

tools/list at https://gateway.pipeworx.io/semanticscholar/mcp returns the tools in the table above plus the shared Pipeworx meta-tools — ask_pipeworx, discover_tools, search_within, remember/recall and the rest of the gateway-wide set. So the tool count you see is larger than this table: a single-pack endpoint currently lists roughly 30 shared tools alongside the pack's own. The connection's initialize response states its exact scope, and is the authoritative answer for a given day.

This is deliberate, not multiplexing by accident. The meta-tools are what let a scoped connection answer a question this pack does not cover — via ask_pipeworx, which routes across the whole catalog — without you adding a second MCP server. There is currently no way to mount a pack endpoint without them; if the extra schemas cost you more context than the routing is worth, connect to the full gateway once rather than to several pack endpoints.

Or connect to the full Pipeworx gateway to get every pack's tools listed directly, instead of just this one's:

{
  "mcpServers": {
    "pipeworx": {
      "url": "https://gateway.pipeworx.io/mcp"
    }
  }
}

Both URLs reach the same gateway and the same 1683+ data sources. The only difference is which pack's tools are listed directly; ask_pipeworx reaches all of them from either one.

No MCP client? Call it over HTTP

curl -X POST https://gateway.pipeworx.io/v1/tools/semanticscholar_search_papers \
  -H 'Content-Type: application/json' \
  -d '{"query":"transformer attention mechanism","year":"2023","limit":10}'

No account needed for the first calls. Inspect any tool: GET https://gateway.pipeworx.io/v1/tools/semanticscholar_search_papers. Find one: POST https://gateway.pipeworx.io/v1/tools/search_packs with {"query":"..."}.

Standalone (no gateway account)

This package also runs as a local stdio MCP server — no Pipeworx account, no gateway round-trip:

{
  "mcpServers": {
    "semanticscholar": {
      "command": "npx",
      "args": ["-y", "@pipeworx/mcp-semanticscholar"]
    }
  }
}

Or run it directly to confirm it starts:

npx -y @pipeworx/mcp-semanticscholar

It speaks MCP over stdin/stdout and answers initialize/tools/list/tools/call for only this pack's tools — none of the shared meta-tools the gateway connection above adds. Same source, same tools, no ask_pipeworx routing.

Using with ask_pipeworx

Instead of calling tools directly, you can ask questions in plain English — this works on the pack endpoint above as well as on the full gateway:

ask_pipeworx({ question: "your question about Semanticscholar data" })

The gateway picks the right tool and fills the arguments automatically.

More

License

MIT

Frequently asked questions

What is Semanticscholar?

Semantic Scholar Academic Graph MCP.

How do I install Semanticscholar?

It runs as a remote server. In Claude Code: claude mcp add --transport http semanticscholar 'https://gateway.pipeworx.io/semanticscholar/mcp'. For other clients, paste the JSON configuration from this page into their MCP settings.

Does Semanticscholar need an API key?

Yes. It expects PIPEWORX_API_KEY, which you set in its configuration.

Is Semanticscholar open source?

Yes. Its code is on GitHub under the MIT license.

Who maintains Semanticscholar?

It is published by pipeworx-io. AI.tn lists it but does not maintain or endorse it.

From the official MCP Registry. Listing is not an endorsement: review a server's code and permissions before giving it access to your data.