Aws Sa Tools MCP Server

🚀 AWS MCP Server

A Model Context Protocol (MCP) server that provides tools for interacting with AWS services.

🚀 Quick Start

The AWS MCP Server is a powerful tool that enables seamless interaction with various AWS services. It simplifies the process of performing common AWS tasks by providing a unified interface.

✨ Features

AWS SDK Tools

The following are the main features of the AWS MCP Server. You can use these features to interact with AWS services. If certain features are not listed here, they can be supported through the aws_sdk_wrapper() tool in aws_mcp_server.tools, provided that the AWS SDK supports the feature and the LLM can generate the commands correctly.

• S3 Operations

  • List buckets
  • List objects in a bucket

• EC2 Operations

  • Describe instances
  • List security groups
  • List VPCs

• RDS Operations

  • Describe database instances

• Cost Explorer

  • Get cost and usage reports
  • Get resource-specific cost analysis

• CloudWatch

  • Get metric statistics

Auxiliary Tools

• Local File System

  • List local folders
  • Read code files

• Terraform

  • Read message status files

📦 Installation

Prerequisites

• Python 3.12 or higher
• Configure AWS credentials (~/.aws/credentials)
• Set up an AWS CLI profile

Installation Requirements

1. Install uv from Astral or GitHub README.
2. Install Python 3.10 or a later version using uv python install 3.10 (or higher).

Configuration

Configure your Claude Desktop settings by adding the following to claude_desktop_config.json:

{
    "mcpServers": {
        "aws-mcp-server": {
            "command": "uvx",
            "args": [
                "--directory",
                "/path/to/aws-mcp-server",
                "."
            ]
        }
    }
}

WSL Configuration

{
    "mcpServers": {
        "aws-mcp-server": {
            "command": "wsl.exe",
            "args": [
                "bash",
                "-c",
                "uvx --directory /path/to/aws-mcp-server ."
            ]
        }
    }
}

Ensure that your AWS credentials are correctly configured in ~/.aws/credentials:

[default]
aws_access_key_id = Your access key
aws_secret_access_key = Your secret key