MCP.so
Sign In

MCP AWS Client

@debedb

About MCP AWS Client

Overview

What is MCP AWS Client?

MCP AWS Client is a Model Context Protocol (MCP) client for AWS services that allows you to manage AWS resources using either AWS CLI or boto3. It is designed to work with any MCP-compatible application, enabling you to automate S3 and Fargate operations within your MCP workflows.

How to use MCP AWS Client?

Clone the repository, install dependencies with pip install -e ., and configure AWS credentials using environment variables (AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY, AWS_REGION), the AWS credential chain (aws configure), or by specifying a profile name when using the tools. After setup, you can use the client with any MCP-compatible application to issue commands for S3 and Fargate operations.

Key features of MCP AWS Client

  • Create, list, and delete S3 buckets
  • Add, list, and remove objects in S3 buckets
  • Modify S3 bucket permissions and policies
  • Deploy new images into AWS Fargate services
  • Supports AWS profiles from ~/.aws/config and ~/.aws/credentials

Use cases of MCP AWS Client

  • Automating S3 bucket and object management as part of an MCP-based pipeline
  • Deploying new container images to AWS Fargate services via MCP commands
  • Updating S3 bucket policies and permissions programmatically through MCP tools

FAQ from MCP AWS Client

Which AWS services does MCP AWS Client support?

The client supports Amazon S3 (bucket and object operations) and AWS Fargate (image deployment into services).

How do I configure AWS credentials?

You can configure credentials using environment variables, the AWS CLI credential chain (aws configure), or by specifying a profile name when using the tools.

Can I use multiple AWS profiles?

Yes, the client reads profiles from ~/.aws/config and ~/.aws/credentials. You can specify a profile name when invoking tools to switch between different AWS accounts or roles.

What are the installation steps?

Clone the repository and run pip install -e . in the project directory. Ensure you have Python installed and your AWS credentials properly set up.

Comments

More Other MCP clients