GnistAI GnistAI
Log in

Getting Started with Semantic Scholar

Search 200M+ academic papers, explore citation graphs, find author profiles, and get AI-powered paper recommendations across all scientific disciplines.

All Tutorials   |   Overview   |   Playground   |   MCP   |   REST API   |   Home
Science

Data source: Semantic Scholar API

Overview

Semantic Scholar wraps Semantic Scholar API, handling authentication, pagination, and rate limits for you. This tutorial covers all 8 tools with working code examples you can copy and run.

Prerequisites

  1. Sign up at https://context.gnist.ai/signup for a free API key (100 calls/day).
  2. Choose your integration method: MCP protocol or REST API.

Connect via MCP

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

MCP Config 
{
  "mcpServers": {
    "gnist-semantic-scholar": {
      "url": "https://context.gnist.ai/mcp/semantic-scholar/",
      "headers": {
        "Gnist-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Tools (8)

search_papers

Search 200M+ academic papers in Semantic Scholar. Find papers across all scientific disciplines. Returns titles, authors, citation counts, venues, and links to open-access PDFs when available. Use paper IDs from results with get_paper, get_citations, get_references, and recommend_papers. Args: query: Search keywords or phrase. limit: Number of results (1-100, default 10). offset: Pagination offset for paging through results. year: Year or year range filter. fields_of_study: Filter by academic field(s). Returns: Papers with paperId, title, year, venue, authors, citationCount, doi.

ParameterTypeRequiredDescription
querystringrequiredSearch query for academic papers (e.g. "transformer attention mechanism", "CRISPR gene editing", "climate change mitigation").
limitintegeroptionalNumber of results to return (1-100, default 10). (default: 10)
offsetintegeroptionalPagination offset for additional results (default 0). (default: 0)
yearanyoptionalYear filter — single year ("2023"), range ("2020-2023"), or open range ("2020-").
fields_of_studyanyoptionalFilter by academic field (e.g. ["Computer Science"], ["Medicine", "Biology"]).
curl -X POST "https://context.gnist.ai/mcp/semantic-scholar/" \
  -H "Content-Type: application/json" \
  -H "Gnist-API-Key: YOUR_API_KEY" \
  -d '{"jsonrpc": "2.0", "method": "tools/call", "id": 1, "params": {"name": "search_papers", "arguments": {"query": "transformer attention mechanism"}}}'
import httpx

resp = httpx.post(
    "https://context.gnist.ai/mcp/semantic-scholar/",
    headers={"Gnist-API-Key": "YOUR_API_KEY"},
    json={'id': 1,
 'jsonrpc': '2.0',
 'method': 'tools/call',
 'params': {'arguments': {'query': 'transformer attention mechanism'},
            'name': 'search_papers'}},
)
print(resp.json())

get_paper

Get detailed information about a specific academic paper. Retrieve full paper details including abstract, citation/reference counts, authors, fields of study, and AI-generated TLDR summary. Accepts multiple ID formats: Semantic Scholar paper ID, DOI, ArXiv ID, PMID, or CorpusId. Args: paper_id: Paper identifier in any supported format. Returns: Full paper details with abstract, authors, citations, TLDR, and links.

ParameterTypeRequiredDescription
paper_idstringrequiredPaper identifier — S2 ID (40-char hex), DOI (e.g. "10.1038/s41586-021-03819-2"), ArXiv ID (e.g. "2301.12345"), or prefixed ID (e.g. "PMID:12345", "CorpusId:12345").
curl -X POST "https://context.gnist.ai/mcp/semantic-scholar/" \
  -H "Content-Type: application/json" \
  -H "Gnist-API-Key: YOUR_API_KEY" \
  -d '{"jsonrpc": "2.0", "method": "tools/call", "id": 1, "params": {"name": "get_paper", "arguments": {"paper_id": "10.1038/s41586-021-03819-2"}}}'
import httpx

resp = httpx.post(
    "https://context.gnist.ai/mcp/semantic-scholar/",
    headers={"Gnist-API-Key": "YOUR_API_KEY"},
    json={'id': 1,
 'jsonrpc': '2.0',
 'method': 'tools/call',
 'params': {'arguments': {'paper_id': '10.1038/s41586-021-03819-2'},
            'name': 'get_paper'}},
)
print(resp.json())

get_citations

Get papers that cite a given paper. Explore a paper's citation graph by finding all papers that reference it. Useful for tracking research impact, finding follow-up work, and discovering how a paper has influenced subsequent research. Args: paper_id: Paper identifier in any supported format. limit: Number of citing papers to return (1-100, default 20). offset: Pagination offset. Returns: List of citing papers with title, year, venue, authors, citationCount.

ParameterTypeRequiredDescription
paper_idstringrequiredPaper identifier (S2 ID, DOI, ArXiv ID, or prefixed ID).
limitintegeroptionalNumber of citing papers to return (1-100, default 20). (default: 20)
offsetintegeroptionalPagination offset (default 0). (default: 0)
curl -X POST "https://context.gnist.ai/mcp/semantic-scholar/" \
  -H "Content-Type: application/json" \
  -H "Gnist-API-Key: YOUR_API_KEY" \
  -d '{"jsonrpc": "2.0", "method": "tools/call", "id": 1, "params": {"name": "get_citations", "arguments": {"paper_id": "12345"}}}'
import httpx

resp = httpx.post(
    "https://context.gnist.ai/mcp/semantic-scholar/",
    headers={"Gnist-API-Key": "YOUR_API_KEY"},
    json={'id': 1,
 'jsonrpc': '2.0',
 'method': 'tools/call',
 'params': {'arguments': {'paper_id': '12345'}, 'name': 'get_citations'}},
)
print(resp.json())

get_references

Get papers referenced by a given paper. Explore a paper's bibliography — the works it cites. Useful for understanding a paper's foundations, finding seminal works in a field, and building reading lists from authoritative sources. Args: paper_id: Paper identifier in any supported format. limit: Number of referenced papers to return (1-100, default 20). offset: Pagination offset. Returns: List of referenced papers with title, year, venue, authors, citationCount.

ParameterTypeRequiredDescription
paper_idstringrequiredPaper identifier (S2 ID, DOI, ArXiv ID, or prefixed ID).
limitintegeroptionalNumber of referenced papers to return (1-100, default 20). (default: 20)
offsetintegeroptionalPagination offset (default 0). (default: 0)
curl -X POST "https://context.gnist.ai/mcp/semantic-scholar/" \
  -H "Content-Type: application/json" \
  -H "Gnist-API-Key: YOUR_API_KEY" \
  -d '{"jsonrpc": "2.0", "method": "tools/call", "id": 1, "params": {"name": "get_references", "arguments": {"paper_id": "12345"}}}'
import httpx

resp = httpx.post(
    "https://context.gnist.ai/mcp/semantic-scholar/",
    headers={"Gnist-API-Key": "YOUR_API_KEY"},
    json={'id': 1,
 'jsonrpc': '2.0',
 'method': 'tools/call',
 'params': {'arguments': {'paper_id': '12345'}, 'name': 'get_references'}},
)
print(resp.json())

search_authors

Search for academic authors by name. Find researchers by name and see their publication counts, citation counts, h-index, and institutional affiliations. Use authorId from results with get_author for detailed profiles. Args: query: Author name to search for. limit: Number of results (1-100, default 10). Returns: Authors with authorId, name, paperCount, citationCount, hIndex, affiliations.

ParameterTypeRequiredDescription
querystringrequiredAuthor name to search for (e.g. "Yann LeCun", "Geoffrey Hinton").
limitintegeroptionalNumber of results to return (1-100, default 10). (default: 10)
curl -X POST "https://context.gnist.ai/mcp/semantic-scholar/" \
  -H "Content-Type: application/json" \
  -H "Gnist-API-Key: YOUR_API_KEY" \
  -d '{"jsonrpc": "2.0", "method": "tools/call", "id": 1, "params": {"name": "search_authors", "arguments": {"query": "Yann LeCun"}}}'
import httpx

resp = httpx.post(
    "https://context.gnist.ai/mcp/semantic-scholar/",
    headers={"Gnist-API-Key": "YOUR_API_KEY"},
    json={'id': 1,
 'jsonrpc': '2.0',
 'method': 'tools/call',
 'params': {'arguments': {'query': 'Yann LeCun'}, 'name': 'search_authors'}},
)
print(resp.json())

get_author

Get an author's profile by Semantic Scholar author ID. Retrieve detailed author information including paper count, citation count, h-index, and affiliations. Args: author_id: Numeric S2 author ID from search_authors results. Returns: Author profile with name, paperCount, citationCount, hIndex, affiliations.

ParameterTypeRequiredDescription
author_idstringrequiredSemantic Scholar author ID (numeric, e.g. "1688882").
curl -X POST "https://context.gnist.ai/mcp/semantic-scholar/" \
  -H "Content-Type: application/json" \
  -H "Gnist-API-Key: YOUR_API_KEY" \
  -d '{"jsonrpc": "2.0", "method": "tools/call", "id": 1, "params": {"name": "get_author", "arguments": {"author_id": "1688882"}}}'
import httpx

resp = httpx.post(
    "https://context.gnist.ai/mcp/semantic-scholar/",
    headers={"Gnist-API-Key": "YOUR_API_KEY"},
    json={'id': 1,
 'jsonrpc': '2.0',
 'method': 'tools/call',
 'params': {'arguments': {'author_id': '1688882'}, 'name': 'get_author'}},
)
print(resp.json())

recommend_papers

Get paper recommendations based on a seed paper. Uses Semantic Scholar's recommendation engine to find related papers you might want to read next. Useful for literature review, discovering adjacent research, and expanding knowledge in a topic area. Args: paper_id: Paper to get recommendations for. limit: Number of recommendations (1-100, default 10). Returns: Recommended papers similar to the seed paper.

ParameterTypeRequiredDescription
paper_idstringrequiredPaper identifier (S2 ID, DOI, ArXiv ID, or prefixed ID) to get recommendations for.
limitintegeroptionalNumber of recommendations (1-100, default 10). (default: 10)
curl -X POST "https://context.gnist.ai/mcp/semantic-scholar/" \
  -H "Content-Type: application/json" \
  -H "Gnist-API-Key: YOUR_API_KEY" \
  -d '{"jsonrpc": "2.0", "method": "tools/call", "id": 1, "params": {"name": "recommend_papers", "arguments": {"paper_id": "12345"}}}'
import httpx

resp = httpx.post(
    "https://context.gnist.ai/mcp/semantic-scholar/",
    headers={"Gnist-API-Key": "YOUR_API_KEY"},
    json={'id': 1,
 'jsonrpc': '2.0',
 'method': 'tools/call',
 'params': {'arguments': {'paper_id': '12345'}, 'name': 'recommend_papers'}},
)
print(resp.json())

report_feedback

Report a bug, feature request, or general feedback for this data source. Use this when something doesn't work as expected, when you'd like a new feature, or when you have suggestions for improvement. Args: feedback: Describe the issue or suggestion. feedback_type: One of 'bug', 'feature_request', or 'general'.

ParameterTypeRequiredDescription
feedbackstringrequired
feedback_typestringoptional (default: general)
curl -X POST "https://context.gnist.ai/mcp/semantic-scholar/" \
  -H "Content-Type: application/json" \
  -H "Gnist-API-Key: YOUR_API_KEY" \
  -d '{"jsonrpc": "2.0", "method": "tools/call", "id": 1, "params": {"name": "report_feedback", "arguments": {"feedback": "example"}}}'
import httpx

resp = httpx.post(
    "https://context.gnist.ai/mcp/semantic-scholar/",
    headers={"Gnist-API-Key": "YOUR_API_KEY"},
    json={'id': 1,
 'jsonrpc': '2.0',
 'method': 'tools/call',
 'params': {'arguments': {'feedback': 'example'}, 'name': 'report_feedback'}},
)
print(resp.json())

Common Patterns

Search then retrieve
Use search_papers to find items, then get_paper to get full details. This two-step pattern is common for exploring data before drilling down.
Pagination
Several tools support limit, offset, or page parameters. Start with small limits during development, then increase for production queries.

FAQ

What data does Semantic Scholar provide?

Search 200M+ academic papers, explore citation graphs, find author profiles, and get AI-powered paper recommendations across all scientific disciplines. It exposes 8 tools: search_papers, get_paper, get_citations, get_references, search_authors, get_author, recommend_papers, report_feedback.

What do I need to get started?

A Gnist API key (free tier: 100 calls/day). Sign up at https://context.gnist.ai/signup.

What format does the Semantic Scholar API return?

JSON, via either MCP protocol (JSON-RPC 2.0) or REST API.

Next Steps

Interactive Playground
Try tools in your browser — no setup required.
Full Reference
Complete docs for all 8 tools.

Related Tutorials

Academic Research
Tutorial
Health Research
Tutorial
Environmental Monitoring
Tutorial
Research Funding
Tutorial
arXiv
Tutorial
ClinicalTrials.gov
Tutorial