AWS Cost Explorer and Amazon Bedrock Model Invocation Logs MCP Server & Client

This project provides an MCP server for accessing AWS spend data via Cost Explorer and Amazon Bedrock usage data via Model Invocation Logs in Amazon CloudWatch, using Anthropic's MCP. It allows users to analyze and visualize AWS cloud spending data using Claude as an interactive interface.

Features: Amazon EC2/Bedrock Spend Analysis, Service Spend Reports, Detailed Cost Breakdown by day, region, service, and instance type, and an Interactive Interface via Claude.

Requirements: Python 3.12, AWS credentials with Cost Explorer access, Anthropic API access, and optional Amazon Bedrock/EC2 access.

Usage: The server can be set up locally or remotely. Local setup uses stdio transport, while remote setup uses sse. The server exposes tools like get_ec2_spend_last_day(), get_detailed_breakdown_by_day(), get_bedrock_daily_usage_stats(), and get_bedrock_hourly_usage_stats(). Example queries include asking about Bedrock spend, EC2 spend, top AWS services by cost, and spending by region. Docker support is included for containerized deployment. A secure remote MCP server setup using nginx as a reverse proxy is also detailed.