Finance Tools

🚀 Finance Tools MCP Server: finance-tools-mcp

The finance-tools-mcp server is based on the Model Context Protocol (MCP), aiming to provide comprehensive financial analysis capabilities for Large Language Models (LLMs).

https://github.com/VoxLink-org/finance-tools-mcp

🚀 Quick Start

The finance-tools-mcp is a server based on the Model Context Protocol (MCP). It is designed to offer comprehensive financial analysis capabilities for Large Language Models (LLMs). Modified from investor-agent, this server integrates multiple data sources and analysis libraries, providing a toolkit to support detailed financial research.

✨ Features

1. Financial Analysis Tools

• Real-time Stock Data Retrieval: Obtain real-time stock prices, trading volumes, and other information from multiple sources.
• Technical Indicator Calculation: Include Moving Average (MA), Relative Strength Index (RSI), MACD, etc.
• Financial Statement Parsing: Parse company financial reports and extract key financial indicators such as revenue, profit, and debt-to-asset ratio.

2. Financial Data Processing

• Data Cleaning and Preprocessing: Denoise and standardize the format of the obtained financial data.
• Time Series Analysis: Use models like ARIMA and LSTM for prediction.
• Risk Management Tools: Include Value at Risk (VaR), credit scoring, etc.

3. Visualization and Reporting

• Interactive Chart Generation: Support various types of financial data visualization, such as candlestick charts, line charts, and bar charts.
• Automatic Report Generation: Automatically generate financial reports and investment suggestions based on the analysis results.

🔧 Technical Details

Time Series Data Processing and Analysis

The server has deeply optimized time series data:

• Efficient Data Storage: Use efficient database technologies to store large-scale time series data.
• Real-time Data Analysis: Support real-time data stream processing with millisecond-level response.
• Prediction Model Integration: Integrate multiple prediction models, such as Facebook Prophet and TensorFlow.

📦 Installation

1. Installation Steps

# Install uvx
sh <https://install.llmops.com/v0.6.0/install.sh>

# Run the finance-tools-mcp server
uvx finance-tools-mcp

2. Environment Variable Configuration

If you have a Federal Reserve Economic Data (FRED) API key, you can configure it as follows:

FRED_API_KEY=YOUR_API_KEY uvx finance-tools-mcp

3. Running Options

• Run using the Server-Sent Events (SSE) transport protocol:

uvx finance-tools-mcp --transport sse

• Combine the FRED API key with the SSE transport:

FRED_API_KEY=YOUR_API_KEY uvx finance-tools-mcp --transport sse

💻 Usage Examples

Integrating with MCP Clients

To integrate finance-tools-mcp into an MCP client (such as Claude Desktop), simply add the following to the configuration file:

{
  "mcpServers": {
    "investor": {
        "command": "path/to/uvx/command/uvx",
        "args": ["finance-tools-mcp"],
    }
  }
}

Debugging

Use the MCP inspection tool for debugging:

npx @modelcontextprotocol/inspector uvx finance-tools-mcp

Or

npx @modelcontextprotocol/inspector uv --directory  ./ run finance-tools-mcp

Log Monitoring Paths

• macOS: ~/Library/Logs/Claude/mcp*.log
• Windows: %APPDATA%\Claude\logs\mcp*.log

Development and Testing

1. Use the MCP inspection tool for debugging.
2. Test in Claude Desktop with the following configuration:

{
  "mcpServers": {
    "investor": {
      "command": "path/to/uv/command/uv",
      "args": ["--directory", "path/to/finance-tools-mcp", "run", "finance-tools-mcp"],
    }
  }
}

📄 License

This MCP server is licensed under the MIT License. Please refer to the LICENSE file for details.

📚 Documentation

Example Reports

• carvana_analysis.md
• palantir_analysis.md
• pdd_analysis_20250503.md
• meli_se_shop_comparison_20250504.md
• GLD_analysis_20250508.md

To-Do List

• [ ] Add support for support and resistance level analysis.
• [x] Added Fibonacci retracement level analysis.
• [ ] Add a Moving Average Convergence Divergence (MACD) prediction model.
• [ ] Develop an options pricing model for prediction.
• [ ] Build a prediction model based on financial statements and other features.