Semantic Scholar MCP Server

What is Semantic Scholar MCP Server?

A FastMCP server implementation for the Semantic Scholar API, providing comprehensive access to academic paper data, author information, and citation networks. It is designed for researchers, developers, and anyone needing programmatic access to scholarly content and citation analysis.

How to use Semantic Scholar MCP Server?

Install the server using FastMCP with the command fastmcp install semantic-scholar-server.py --name "Semantic Scholar" -e SEMANTIC_SCHOLAR_API_KEY=your-api-key. The API key is optional; without it the server uses unauthenticated access with lower rate limits. After installation, tools are available via any MCP client.

Key features of Semantic Scholar MCP Server

• Full-text search with advanced filtering and ranking
• Citation network exploration and reference tracking
• Author search, profile details, and publication history
• Paper recommendations (single and multi-paper)
• Batch retrieval for up to 1000 papers or authors
• Rate limiting compliance for authenticated and unauthenticated access

Use cases of Semantic Scholar MCP Server

• Discover highly cited papers in a specific field using relevance search
• Build citation graphs for a paper by exploring its citations and references
• Get profile metrics (h-index, citation count) for multiple authors at once
• Receive paper recommendations based on one or more seed papers
• Bulk fetch metadata for dozens of papers to populate a literature review

FAQ from Semantic Scholar MCP Server

Do I need an API key to use Semantic Scholar MCP Server?

No. If you do not provide a SEMANTIC_SCHOLAR_API_KEY, the server uses unauthenticated access with lower rate limits.

What are the rate limits for Semantic Scholar MCP Server?

With an API key: search, batch and recommendation endpoints are limited to 1 request per second; other endpoints to 10 requests per second. Without an API key: all endpoints are limited to 100 requests per 5 minutes.

What paper ID formats does Semantic Scholar MCP Server accept?

It accepts various formats including S2 ID, DOI, and ArXiv identifiers.

How many papers or authors can I retrieve in a single batch request?

Batch endpoints (paper_batch_details, author_batch_details) accept up to 1000 IDs per request.

How does Semantic Scholar MCP Server handle errors?

The server returns standardized error responses containing an error type (rate_limit, api_error, validation, timeout), a human-readable message, and additional context such as whether the request was authenticated.