🏦 FinSight MCP Server
A production-grade Model Context Protocol (MCP) Server built with Java 21 + Spring Boot 3, exposing real-time financial tools to AI assistants like Claude. This project demonstrates how to build an MCP server that bridges AI models with live financial market data.
🎯 What is MCP?
Model Context Protocol (MCP) is an open standard that allows AI assistants to call external tools and access real-time data. Instead of relying on training data, the AI can invoke your server's tools to get live information.
Claude (AI Client)
│ MCP Protocol (SSE)
▼
FinSight MCP Server (Spring Boot 3 + Java 21)
│
├── Alpha Vantage API (stock quotes, fundamentals)
└── NewsAPI (financial news)
🔧 Tools Exposed
| Tool | Description |
|---|---|
| getStockQuote | Fetch real-time stock price for a single ticker |
| compareStocks | Side-by-side comparison of multiple tickers |
| analyzePortfolio | Portfolio breakdown with dollar allocation per holding |
| getFinancialNews | Latest news articles for a given stock ticker |
| screenStocks | Filter stocks from a predefined list by financial criteria |
🛠️ Tech Stack
| Layer | Technology |
|---|---|
| Language | Java 21 |
| Framework | Spring Boot 4.0.3 |
| MCP SDK | Spring AI MCP Server (WebFlux) |
| HTTP Client | WebClient (Reactive) |
| Transport | SSE (Server-Sent Events) |
| Runtime | Netty |
| Market Data | Alpha Vantage API |
| News Data | NewsAPI |
| Container | Docker |
📁 Project Structure
finsight-mcp-server/
├── src/main/java/com/finsight/
│ ├── config/
│ │ └── McpConfig.java # Bean registration for MCP tools + WebClients
│ ├── model/
│ │ ├── GlobalQuote.java # Alpha Vantage quote response model
│ │ ├── StockQuoteResponse.java # Wrapper for quote API response
│ │ ├── StockOverview.java # Alpha Vantage overview response model
│ │ ├── PortfolioRequest.java # Portfolio analysis input model
│ │ ├── Stock.java # Individual stock holding model
│ │ ├── ScreenRequest.java # Stock screener filter criteria
│ │ ├── NewsResponse.java # NewsAPI response wrapper
│ │ ├── Article.java # Individual news article model
│ │ └── Source.java # News source model
│ ├── service/
│ │ ├── AlphaVantageService.java # Alpha Vantage API integration
│ │ └── NewsService.java # NewsAPI integration
│ └── tools/
│ └── StockTool.java # MCP tool definitions
├── src/main/resources/
│ └── application.properties # App configuration
├── Dockerfile
└── pom.xml
🚀 Getting Started
Prerequisites
Run Locally
# Clone the repository
git clone https://github.com/YOUR_USERNAME/finsight-mcp-server.git
cd finsight-mcp-server
# Set environment variables
export APP_STOCK_API=your_alpha_vantage_key
export APP_NEWS_API_KEY=your_news_api_key
# Build and run
mvn clean package -DskipTests
java -jar target/finsight-mcp-server-0.0.1-SNAPSHOT.jar
Run with Docker
# Build image
docker build -t finsight-mcp-server .
# Run container
docker run -p 8080:8080 \
-e APP_STOCK_API=your_alpha_vantage_key \
-e APP_NEWS_API_KEY=your_news_api_key \
finsight-mcp-server
🔌 Connect to Claude Desktop
Add this to your claude_desktop_config.json:
{
"mcpServers": {
"finsight": {
"command": "npx",
"args": [
"mcp-remote",
"http://localhost:8080/sse"
]
}
}
}
For the deployed version, replace http://localhost:8080 with your Railway URL.
💬 Example Prompts
Once connected to Claude Desktop, try:
What is the current stock price of AAPL?
Compare the s
... [View full README on GitHub](https://github.com/shubhamkhandelwal/finsight-mcp-server#readme)
