AWS MCP Server - Model Context Protocol Server

About this MCP

AWS MCP Server provides a secure bridge between MCP-aware AI assistants (like Claude Desktop, Cursor, and Windsurf) and the AWS Command Line Interface. It allows AI tools to retrieve AWS CLI documentation and execute AWS commands in a containerized environment, returning formatted results optimized for AI consumption. This service enhances cloud infrastructure management by enabling AI assistants to directly interact with AWS resources through natural language requests. With support for Unix pipe operations and pre-defined prompt templates for common AWS tasks, it streamlines complex cloud operations while maintaining security through containerization.

Documentation

Overview

AWS MCP Server implements the Model Context Protocol (MCP) to create a secure interface between AI assistants and the AWS Command Line Interface. This allows AI tools like Claude, Cursor, and other MCP-aware assistants to help you manage your AWS infrastructure more effectively.

Features

AWS CLI Documentation Access: Retrieve detailed help information for AWS CLI commands
Secure Command Execution: Run AWS CLI commands in a containerized environment
Unix Pipe Support: Filter and transform AWS CLI output using standard Unix pipes and utilities
Prompt Templates: Pre-defined templates for common AWS tasks following best practices
Multi-architecture Support: Works on both AMD64/x86_64 and ARM64 platforms

Installation

Option 1: Using Docker (Recommended)

# Clone the repository
git clone https://github.com/alexei-led/aws-mcp-server.git
cd aws-mcp-server

# Build and run Docker container
docker compose -f deploy/docker/docker-compose.yml up -d

You can also pull the pre-built multi-architecture image:

# Latest stable version
docker pull ghcr.io/alexei-led/aws-mcp-server:latest

# Or specific version (recommended for production)
docker pull ghcr.io/alexei-led/aws-mcp-server:1.0.0

Option 2: Using Python

Requirements:

Python 3.13+
AWS CLI installed locally
AWS credentials configured

# Clone repository
git clone https://github.com/alexei-led/aws-mcp-server.git
cd aws-mcp-server

# Set up virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install in development mode
pip install -e .

# Run the server
python -m aws_mcp_server

Configuration

The server uses your existing AWS credentials. Make sure you have configured AWS CLI with appropriate credentials before running the server.

Usage with AI Assistants

Start the AWS MCP Server using one of the installation methods above
Open your MCP-aware AI assistant (Claude Desktop, Cursor, etc.)
Connect to the MCP server (typically at http://localhost:8080)
Ask your AI assistant to help with AWS tasks, such as:
- "Show me my EC2 instances"
- "Create a new S3 bucket with versioning enabled"
- "Help me set up an ECS cluster"

The AI assistant will use the MCP server to execute the necessary AWS CLI commands and provide you with the results.

Security Considerations

The server runs AWS CLI commands in a containerized environment for added security
It uses your existing AWS credentials, so commands will only have the permissions you've already configured
For production use, consider implementing additional authentication for the MCP server itself

Troubleshooting

If you encounter issues:

Check that your AWS credentials are properly configured
Verify the server is running (docker ps or check process list)
Ensure your AI assistant is correctly configured to use the MCP server
Check the server logs for any error messages

For more detailed information, visit the GitHub repository.