scholar-mcp
Multi-source academic paper search, citation graph exploration, and PDF download as an MCP server, designed for LLM agents doing research.
README
scholar-mcp
Multi-source academic paper search, citation graph exploration, and PDF download as an MCP server. Designed for LLM agents doing research.
Fuses results from 9 academic sources via Reciprocal Rank Fusion, with PDF access across 10+ preprint servers.
Quick Start
Claude Code:
claude mcp add scholar -- uvx scholar-mcp
With API key (recommended, higher rate limits):
claude mcp add scholar -e S2_API_KEY=your_key -- uvx scholar-mcp
Claude Desktop (add to config):
{
"mcpServers": {
"scholar": {
"command": "uvx",
"args": ["scholar-mcp"]
}
}
}
Requires Python 3.10+ and uv. No API key needed for basic use.
Tools
| Tool | Description |
|---|---|
search_papers |
Multi-source search with RRF fusion. Filters: year, venue, field, citations, open access |
get_paper |
Paper details by S2 ID, DOI, ArXiv ID, PMID, or OpenAlex ID |
get_citations |
Papers citing a given paper (impact-sorted) |
get_references |
Papers referenced by a given paper |
recommend_papers |
Similar papers via SPECTER2 embeddings |
search_authors |
Researchers with h-index, affiliations, paper counts |
build_paper_graph |
Citation graph with PageRank analytics and Mermaid visualization |
search_openreview |
Conference papers (ICLR, NeurIPS, ICML) |
discover_field |
Auto-map a research field: find surveys, foundations, recent trends, build citation graph |
save_papers |
Save papers to a persistent collection for later reference |
list_saved_papers |
List or search saved paper collections (persists across sessions) |
download_paper |
Smart PDF download across 10+ sources |
read_paper |
Download + extract text from PDF |
Search Sources
| Source | Coverage | Strength |
|---|---|---|
| Semantic Scholar | 214M papers | SPECTER2 semantic search |
| OpenAlex | 250M works | Best coverage, impact-ranked citations |
| arXiv | CS/Math/Physics | Preprints |
| PubMed | 36M biomedical | Medicine, biology |
| Europe PMC | Biomedical + EU | PubMed superset |
| Crossref | 150M DOIs | Metadata |
| DBLP | CS bibliography | Conferences, proceedings |
| INSPIRE-HEP | High-energy physics | Particle physics |
| CORE | 250M open access | Institutional repositories |
PDF Download Chain
- Semantic Scholar open access
- arXiv direct
- CORE (institutional repositories)
- Preprint servers: bioRxiv, medRxiv, SSRN, ChemRxiv, PsyArXiv, EarthArXiv, SocArXiv, engrXiv, AgriXiv, SportRxiv, Preprints.org
- Unpaywall (legal OA discovery)
- PubMed Central
- Sci-Hub (opt-in via
SCIHUB_ENABLED=1)
Citation Graph
build_paper_graph builds an interactive citation network:
- BFS expansion with velocity-weighted priority (new + influential papers first)
- PageRank and betweenness centrality via networkx
- Pivot/bridge paper detection
- Topic filtering to keep graph focused
- Mermaid output with color-coded nodes (seed, high-cite, bridge)
build_paper_graph("Attention Is All You Need", max_hops=2, max_papers=20, topic_filter="attention transformer")
Configuration
| Variable | Default | Description |
|---|---|---|
S2_API_KEY |
- | Semantic Scholar API key (1 req/s) |
CORE_API_KEY |
- | CORE API key |
OPENALEX_EMAIL |
- | Email for Unpaywall + OpenAlex polite pool |
SCHOLAR_DOWNLOAD_DIR |
./downloads |
PDF save directory |
SCIHUB_ENABLED |
false |
Enable Sci-Hub as last-resort source |
Development
git clone https://github.com/Liyux3/scholar-mcp.git
cd scholar-mcp
uv venv && uv pip install -e ".[dev]"
uv run pytest tests/ # 42 unit tests
S2_API_KEY=... uv run python tests/smoke_test.py # quick integration check
42 tests: relevance scoring (26), graph analytics (9), knowledge base (7).
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.