Semantic Scholar MCP Server

Semantic Scholar MCP Server

Enables searching and retrieving academic papers, authors, citations, and references from Semantic Scholar via an HTTP MCP server with caching and rate limiting.

Category
访问服务器

README

Semantic Scholar MCP Server

Remote MCP server with Streamable HTTP transport for Semantic Scholar API access.

Features

  • 🔌 Streamable HTTP Transport - Remote MCP server accessible anywhere
  • 💾 SQLite Caching - Persistent cache with TTL-based expiration
  • 🚦 Rate Limiting - Token bucket algorithm to prevent API quota exhaustion
  • 📊 Metrics - Prometheus-compatible metrics endpoint
  • 📝 Structured Logging - JSON logs with request IDs
  • 🔒 Security - API key via environment variable only (no CLI exposure)

Quick Start

Installation

# Clone the repository
git clone https://github.com/your-org/semantic-scholar-mcp.git
cd semantic-scholar-mcp

# Install dependencies
npm install

# Build TypeScript
npm run build

Configuration

Set environment variables:

# Required: Semantic Scholar API key (get from https://www.semanticscholar.org/product/api)
export SEMANTIC_SCHOLAR_API_KEY="your-api-key"

# Optional: Server configuration
export PORT=3000
export HOST="0.0.0.0"
export LOG_LEVEL="info"

# Optional: Cache configuration
export CACHE_TTL_DAYS=7
export CACHE_MAX_SIZE=1073741824

# Optional: Rate limiting
export RATE_LIMIT_REQUESTS_PER_MINUTE=10
export RATE_LIMIT_BURST_SIZE=5

Running

# Production
npm start

# Development (with hot reload)
npm run dev

Authentication

Simple API Key (Default)

Uses environment variable for authentication. Good for personal use.

export SEMANTIC_SCHOLAR_API_KEY="your-api-key"

OAuth 2.1 (Enterprise)

For production multi-user deployments, use the OAuth 2.1 pattern:

Reusable OAuth Pattern: https://github.com/truaxki/mcp-oauth-pattern

The OAuth pattern provides:

  • OAuth 2.1 authentication with Supabase
  • Bearer token middleware
  • Consent page for user authorization
  • Per-user rate limiting (optional)

To add OAuth to this server:

  1. Clone https://github.com/truaxki/mcp-oauth-pattern
  2. Copy src/auth.ts, src/logger.ts from OAuth pattern
  3. Replace API key auth with Bearer token middleware
  4. Configure Supabase environment variables

Connect from Claude Code

Add to your .mcp.json:

{
  "mcpServers": {
    "semantic-scholar": {
      "type": "streamable-http",
      "url": "http://your-server:3000"
    }
  }
}

Endpoints

Endpoint Description
POST /mcp MCP protocol handler
GET /health Health check
GET /metrics Prometheus metrics

Available Tools

Tool Description
search_papers Search for papers with filters
get_paper Get paper details by ID
get_authors Get authors for a paper
get_citations Get papers that cite a paper
get_references Get papers referenced by a paper
batch_fetch Fetch multiple papers efficiently
get_cache_stats View cache statistics
clear_cache Clear cached data

Deployment

Docker

# Build image
docker build -t semantic-scholar-mcp .

# Run container
docker run -d \
  --name semantic-scholar-mcp \
  -p 3000:3000 \
  -e SEMANTIC_SCHOLAR_API_KEY="your-key" \
  semantic-scholar-mcp

Railway Deployment

One-click deploy:

Deploy on Railway

Manual deployment:

  1. Connect repository

    • Go to Railway and sign in
    • Click "New Project" → "Deploy from GitHub repo"
    • Select truaxki/semantic-scholar-mcp
  2. Configure environment variables

    • In Railway dashboard, go to your service's "Variables" tab
    • Add:
      SEMANTIC_SCHOLAR_API_KEY=your-api-key-here
      LOG_LEVEL=info
      PORT=3000
      
  3. Deploy

    • Railway automatically detects Node.js and builds
    • Deploys with Streamable HTTP on port 3000
    • Gets free HTTPS endpoint automatically
  4. Connect from Claude Code

    {
      "mcpServers": {
        "semantic-scholar": {
          "type": "streamable-http",
          "url": "https://your-service.up.railway.app/mcp"
        }
      }
    }
    

Railway pricing: Free tier available, $5/month for dedicated resources.

Manual Server (srv1338041)

# SSH to server
ssh user@srv1338041

# Clone and build
git clone your-repo
cd semantic-scholar-mcp
npm install
npm run build

# Create systemd service (see deploy/semantic-scholar-mcp.service)
sudo cp deploy/semantic-scholar-mcp.service /etc/systemd/system/
sudo systemctl daemon-reload
sudo systemctl enable semantic-scholar-mcp
sudo systemctl start semantic-scholar-mcp

Architecture

┌─────────────────────────────────────────────────────┐
│              Semantic Scholar MCP Server             │
├─────────────────────────────────────────────────────┤
│                                                      │
│  ┌──────────────┐  ┌──────────────┐  ┌────────────┐ │
│  │   Express   │──▶│   MCP SDK   │──▶│  Tools    │ │
│  │   HTTP      │  │   Server    │  │  Registry │ │
│  └──────────────┘  └──────────────┘  └────────────┘ │
│         │                 │                           │
│         ▼                 ▼                           │
│  ┌──────────────┐  ┌──────────────┐                  │
│  │   Logging   │  │   Cache     │                  │
│  │  (Pino)     │  │ (SQLite)    │                  │
│  └──────────────┘  └──────────────┘                  │
│                                                      │
└─────────────────────────────────────────────────────┘
         │                 │                           │
         ▼                 ▼                           ▼
  HTTPS Endpoint    Persistent Cache    Rate Limited API

API Key

Get your free Semantic Scholar API key from: https://www.semanticscholar.org/product/api

Free tier: 100 requests/day

License

MIT

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选