MCP-AWS logo

MCP-AWS

by anirban1592

MCP-AWS is an AI-driven application that uses OpenAI Agents and MCP servers to manage AWS EC2 instances. It allows users to provision and terminate EC2 instances using natural language commands.

View on GitHub

Last updated: N/A

What is MCP-AWS?

MCP-AWS is a simple application that integrates OpenAI Agents with a custom MCP server to manage AWS EC2 instances using natural language commands. It provides tools to create and terminate EC2 instances through AI interaction.

How to use MCP-AWS?

To use MCP-AWS, you need to clone the repository, configure your AWS credentials and OpenAI API key in a .env file, and then either build and run the Docker image or set up a local Python environment. Once running, you can interact with the AI agent through the terminal by providing commands like 'Create an EC2 instance' or 'Terminate EC2 instance with ID <instance-id>'.

Key features of MCP-AWS

  • Provision EC2 Instances via natural language

  • Terminate EC2 Instances via natural language

  • MCP Server Integration with OpenAI Agents SDK

  • Customizable AWS resource management

Use cases of MCP-AWS

  • Automated EC2 instance provisioning

  • Automated EC2 instance termination

  • Learning how to integrate MCP servers with OpenAI Agents

  • Building AI-driven applications for AWS resource management

FAQ from MCP-AWS

What is an MCP server?

An MCP server is a custom server that provides tools for specific tasks, in this case, managing AWS EC2 instances.

What are the prerequisites for running MCP-AWS?

You need Python 3.12+ or Docker, an AWS IAM role with EC2 management permissions, and environment variables for AWS credentials and OpenAI API key.

How do I create an IAM role for MCP-AWS?

You need to create an IAM role with permissions to create and terminate EC2 instances. Follow AWS best practices for security when creating the role.

Is MCP-AWS production-ready?

No, MCP-AWS is a proof of concept (POC) and is intended for learning purposes only. It is not recommended for production use without thorough security and billing considerations.

What if I encounter billing issues or security incidents?

The creators of MCP-AWS are not responsible for any billing issues or security incidents. Users are responsible for their own AWS resource management and security practices.