AWS Resources MCP Server

This project provides a Model Context Protocol (MCP) server that allows querying AWS resources using boto3 through generated Python code. It's designed to offer more flexibility and control compared to services like AWS Chatbot.

Overview: The server exposes an aws://query_resources resource and a query_aws_resources tool. The tool executes Python code snippets, enabling users to interact with AWS services. It requires a code_snippet input and sets a result variable with the query output. Allowed imports include boto3, operator, json, datetime, and pytz.

Setup: The server can be installed via Smithery or Docker. Docker installation involves pulling the image or building it locally, requiring AWS credentials via environment variables (AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY, AWS_DEFAULT_REGION) or an AWS profile.

Usage: The server is intended for use with Claude Desktop, with configuration examples provided for both Docker and Git clone setups.