Frequently Asked Questions (FAQ) about AWS MCP

Q: What is AWS MCP?

A: AWS MCP (Model Context Protocol) is a server that enables AI assistants like Claude to interact with your AWS environment using natural language. It allows you to query and manage your AWS resources through conversational commands.

Q: What are the prerequisites for using AWS MCP?

A: You need Python, Claude Desktop, and properly configured AWS credentials on your local machine.

Q: How do I install AWS MCP?

A: You can install AWS MCP by cloning the GitHub repository, installing the Python package using pip, and configuring Claude Desktop to communicate with the MCP server.

Q: Can I use AWS MCP without Claude Desktop?

A: Yes, you can use AWS MCP as a command-line interface (CLI) to execute AWS operations directly from your terminal.

Q: What AWS services are supported by AWS MCP?

A: AWS MCP dynamically recognizes and works with all AWS services available through boto3, including S3, EC2, Lambda, IAM, DynamoDB, RDS, and many more.

Q: How does AWS MCP handle errors?

A: AWS MCP handles errors gracefully and provides detailed logs that can help you diagnose and resolve any issues.

Q: Where can I find the logs for AWS MCP?

A: You can find the logs in ~/Library/Logs/Claude/mcp-server-aws-mcp.log or ~/Library/Logs/Claude/mcp.log on macOS.

Q: Is AWS MCP secure?

A: AWS MCP uses your existing AWS credentials and executes locally, ensuring security and control over your data.

Q: What is the Model Context Protocol (MCP)?

A: MCP is an open protocol that standardizes how applications provide context to Large Language Models (LLMs).

Q: How does AWS MCP relate to UBOS?

A: AWS MCP is a specialized tool for managing AWS with AI, while UBOS is a broader platform for developing and deploying AI agents for various business needs. They can be used together to manage your cloud infrastructure and automate your business processes.