rickyb30/pulse-mcp-server

Pulse 🔍⚡

A comprehensive Model Context Protocol (MCP) server built with FastMCP 2.0 that provides weather data, web search, calculations, AWS cost analysis, stock market analysis, and Snowflake cost monitoring capabilities.

🌟 Features

Core Tools

• 🧮 Calculator: Statistical operations (sum, average, min, max)
• 🌤️ Weather Data: Real-time weather information using OpenWeather API
• 🔍 Web Search: DuckDuckGo search with OpenAI fallback
• 🏦 AWS Cost Analysis: Multi-account AWS cost monitoring and optimization
• 📈 Stock Market Analysis: Real-time prices, technical indicators, portfolio tracking
• ❄️ Snowflake Cost Monitoring: Warehouse costs, credit usage, storage analysis

AWS Cost Analysis Features

• 📊 Profile Discovery: Automatically discovers AWS profiles from ~/.aws/config and ~/.aws/credentials
• 💰 Cost Analysis: Analyzes running costs across multiple AWS accounts
• 📈 Top 5 Services: Identifies the most expensive AWS services by cost
• 📋 Detailed Reports: Comprehensive cost breakdowns with percentages
• 🎯 Multi-Account Support: Aggregates costs across all configured AWS profiles

Stock Market Analysis Features

• 📊 Real-time Data: Current prices, changes, volume for any stock
• 🔍 Technical Analysis: RSI, MACD, moving averages, Bollinger Bands
• 💼 Portfolio Tracking: Performance analysis, gains/losses, allocations
• 🎯 Stock Screening: Filter stocks by P/E ratio, market cap, price ranges
• 💰 Crypto Support: Bitcoin, Ethereum, and major cryptocurrencies
• 📈 Market Indices: S&P 500, NASDAQ, Dow Jones, VIX tracking

Snowflake Cost Analysis Features

• 🔐 SSO Authentication: External browser authentication for company SSO
• 💰 Overall Cost Monitoring: Compute and storage cost breakdown
• 🏭 Warehouse Analysis: Top 5 warehouses by cost and credit usage
• 📊 Credit Usage Tracking: Detailed credit consumption analysis
• 💾 Storage Cost Analysis: Data, stage, and failsafe storage costs

Resources & Prompts

• Server Status: Real-time server information
• Configuration: Server settings and capabilities
• AWS Profiles: Discovered AWS profile information
• Market Overview: Current market indices and crypto prices
• Analysis Templates: Pre-built prompts for AWS, stock, and Snowflake analysis

🚀 Quick Start

Prerequisites

• Python 3.11+
• Virtual environment
• AWS CLI configured (for AWS cost analysis)
• API Keys (optional):
  • OpenWeather API key
  • OpenAI API key

Installation

1. Clone and Setup:

git clone https://github.com/yourusername/pulse-mcp-server.git
cd pulse-mcp-server
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
pip install -r requirements.txt

2. Configure Environment (optional):

python setup_env.py

3. Configure AWS (for cost analysis):

aws configure  # Follow prompts to set up your AWS credentials

🔧 Usage

Option 1: HTTP Server (Standalone)

# Start HTTP server on localhost:8000
python run_http_server.py

Option 2: Claude Desktop Integration

1. Copy configuration from claude_config.json to:

   • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
   • Windows: %APPDATA%\Claude\claude_desktop_config.json

2. Update paths in the configuration file

3. Restart Claude Desktop

4. Look for the tools icon (🔨)

Option 3: VSCode Integration

1. Built-in MCP Support: Enable chat.mcp.enabled in VSCode settings
2. Open the project in VSCode (.vscode/mcp.json is pre-configured)
3. Use Chat view (Ctrl+Alt+I) in Agent mode
4. Alternative: Install Cline extension for additional MCP features

📖

🏦 AWS Cost Analysis

Available AWS Tools

1. Discover AWS Profiles

discover_aws_profiles()

• Scans ~/.aws/config and ~/.aws/credentials
• Returns all configured AWS profiles
• Shows profile details and credentials status

2. Analyze AWS Costs

analyze_aws_costs(days=30, profile=None)

• days: Number of days to analyze (default: 30)
• profile: Specific profile to analyze (if None, analyzes all)
• Returns detailed cost breakdown by service
• Identifies top 5 most expensive services

3. Get Cost Report

get_aws_cost_report(days=30, format_type="detailed")

• format_type: "detailed" or "summary"
• Returns formatted text report
• Includes cost optimization insights

Example Usage with Claude Desktop

Try these commands in Claude Desktop:

AWS Cost Analysis:

• "Analyze my AWS costs for the last 30 days"
• "Show me the top 5 most expensive AWS services"
• "Generate a cost report for my AWS accounts"
• "Which AWS profile is costing me the most?"

Stock Market Analysis:

• "Get Apple stock information and technical analysis"
• "Show me the current market indices"
• "Analyze my portfolio with AAPL 100 shares at $150 cost basis"
• "Screen stocks with P/E ratio under 25"
• "What's Bitcoin's current price?"

Snowflake Cost Analysis:

• "Connect to Snowflake using SSO with account xyz123 and user john.doe"
• "Show me overall Snowflake costs for the last 30 days"
• "Which are my top 5 most expensive warehouses?"
• "Generate a Snowflake cost report"

Other Tools:

• "What's the weather in Tokyo?"
• "Calculate the average of 10, 20, 30, 40, 50"
• "Search for information about Python programming"

📊 Sample AWS Cost Report

🏦 AWS COST ANALYSIS REPORT
==================================================
📅 Analysis Date: 2025-06-19T21:47:50
📊 Period: Last 30 days
👥 Total Profiles: 2
✅ Successful: 2
❌ Failed: 0
💰 Total Cost (All Accounts): $127.45

🌟 TOP 5 SERVICES (ALL ACCOUNTS)
----------------------------------------
1. Amazon EC2
   💰 $89.32 (70.1%)
2. Amazon RDS
   💰 $23.45 (18.4%)
3. Amazon S3
   💰 $8.92 (7.0%)
4. Amazon CloudWatch
   💰 $3.21 (2.5%)
5. AWS Lambda
   💰 $2.55 (2.0%)

💸 MOST EXPENSIVE ACCOUNT
------------------------------
Profile: production
Account ID: 123456789012
Cost: $98.76

🛠️ Configuration

Environment Variables

# API Keys (optional)
OPENWEATHER_API_KEY=your_openweather_key
OPENAI_API_KEY=your_openai_key

# Server Configuration
MCP_TRANSPORT=stdio  # or http
MCP_HOST=localhost
MCP_PORT=8000

AWS Configuration

The server automatically discovers AWS profiles from:

• ~/.aws/config - AWS configuration file
• ~/.aws/credentials - AWS credentials file

Ensure your AWS credentials have the following permissions:

• ce:GetCostAndUsage (Cost Explorer)
• sts:GetCallerIdentity (Account information)

📁 Project Structure

pulse-mcp-server/
├── .gitignore                  # Git ignore rules
├── .vscode/                    # VSCode configuration
│   ├── mcp.json                # Official VSCode MCP config
│   ├── settings.json           # Workspace settings
│   ├── launch.json             # Debug configurations
│   └── extensions.json         # Recommended extensions
├── LICENSE                     # MIT license
├── README.md                   # Comprehensive documentation
├── VSCODE_SETUP.md             # VSCode integration guide
├── advanced_mcp_server.py      # Main MCP server with all tools
├── aws_cost_analyzer.py        # AWS cost analysis engine
├── stock_market_analyzer.py    # Stock market analysis engine
├── snowflake_cost_analyzer.py  # Snowflake cost monitoring engine
├── run_http_server.py          # Flexible server runner
├── claude_config.json          # Claude Desktop configuration
├── requirements.txt            # Python dependencies
└── setup_env.py               # Environment setup utility

🔍 Troubleshooting

AWS Issues

• No profiles found: Run aws configure to set up credentials
• Permission denied: Ensure your AWS user has Cost Explorer permissions
• Region errors: Cost Explorer API requires us-east-1 region

General Issues

• Import errors: Ensure virtual environment is activated
• API key errors: Check .env file configuration
• Server startup: Verify all dependencies are installed

🚀 Advanced Usage

Custom Cost Analysis

from aws_cost_analyzer import AWSCostAnalyzer

analyzer = AWSCostAnalyzer()
result = analyzer.analyze_all_profiles(days=7)  # Last 7 days
print(result)

HTTP API Integration

# Start HTTP server
export MCP_TRANSPORT=http
python run_http_server.py

# Test AWS cost analysis via HTTP
curl -X POST http://localhost:8000 \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"analyze_aws_costs","arguments":{"days":30}}}'

📚 Documentation

• FastMCP Documentation: Official FastMCP docs
• AWS Cost Explorer API: AWS Cost Explorer documentation
• Yahoo Finance API: Stock market data source
• Snowflake Python Connector: Snowflake integration

🤝 Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests for new functionality
5. Submit a pull request

📄 License

This project is licensed under the MIT License.

🙏 Acknowledgments

• FastMCP: Excellent MCP framework
• AWS SDK: Boto3 for AWS integration
• OpenWeather: Weather data API
• DuckDuckGo: Web search API