stock_mcp_server

This MCP server is designed to help Claude analyze your Indian stock portfolio by connecting to a local MongoDB data store, querying stock data directly, and leveraging Alpha Vantage for NSE/BSE market data within the free tier limits. It also maintains a persistent knowledge graph of your stock analyses to improve context over time. Once running, Claude can access tools to summarize portfolios, analyze segments of holdings, fetch stock-specific metrics, and generate tailored recommendations for NSE/BSE stocks. The server segments large portfolios into smaller analyses to avoid timeouts and to provide quicker, actionable insights.

Prerequisites:

• Python 3.9+
• MongoDB running on localhost:27017 with database "stock_data"
• Alpha Vantage API key (free tier supported)
• Claude Desktop application

Setup steps:

1. Clone the repository and navigate to the project folder:

git clone <repository-url>
cd stock-mcp-server

2. Create and activate a Python virtual environment (optional but recommended):

python -m venv venv
source venv/bin/activate  # on macOS/Linux
venv\Scripts\activate     # on Windows

3. Install dependencies:

pip install -r requirements.txt

4. Create a .env file from the example and configure settings:

cp .env.example .env

Edit .env to configure:

• MONGODB_URI=mongodb://localhost:27017
• MONGODB_DB_NAME=stock_data
• ALPHA_VANTAGE_API_KEY=your_api_key_here
• ALPHA_VANTAGE_DEFAULT_EXCHANGE=NSE # NSE or BSE

5) Start the MCP server:
```bash
python server.py

6. Configure Claude Desktop to connect to this MCP server (as shown in the README example). Ensure the server name in Claude config matches MCP_SERVER_NAME and that all paths are absolute.

Tips and common issues:

• Ensure MongoDB is running and accessible at the configured URI. If connection issues arise, verify MONGODB_URI and MongoDB service status.
• For Claude tool availability, after starting the server, restart Claude or open a new conversation to load the new tools.
• The server uses a segmented analysis approach for large portfolios; you can explicitly request segments and control segment size with the segment_size parameter.
• If you encounter Python path issues in Claude configuration, provide the full path to the Python executable and the full path to server.py.
• Manage Alpha Vantage rate limits by relying on the built-in limiter; for larger analyses, consider segmenting requests to stay within free-tier quotas.
• Environment variables in the Claude config must exactly match the MCP configuration (MCP_SERVER_NAME should match the key value).
• Review logs at Claude logs folder (for example, ~/Library/Logs/Claude/mcp-server-stock-analysis-mcp.log) for troubleshooting tool loading or connection problems.