Would like to know AWS service to support MCP now a day

Great question! AWS now has excellent day-1 MCP support through a couple of key servers:

AWS API MCP Server - This is what you'll want for comprehensive AWS integration. It provides access to ALL AWS services through boto3 operations and CLI commands, including the latest APIs that may have been released after your AI model's knowledge cutoff. The server ships with two tools, call_aws and suggest_aws_commands - the suggest tool is particularly helpful for discovering new AWS CLI commands based on natural language queries.

AWS Knowledge MCP Server - Complements the API server by providing access to AWS documentation and best practices. It's a fully managed remote server that gives you real-time access to AWS docs, API references, What's New announcements, Well-Architected guidance, and architectural patterns - no AWS account required and less local setup since it's hosted. It has tools like search_documentation and read_documentation.

I'd recommend starting with both of these - use the Knowledge server for guidance and the API server for actual AWS operations. They work great with clients like Amazon Q CLI, Cursor, and Cline.

Installation instructions: https://github.com/awslabs/mcp/tree/main/src/aws-knowledge-mcp-server https://github.com/awslabs/mcp/tree/main/src/aws-api-mcp-server

Besides these, there are several more specialized AWS MCP servers here: https://github.com/awslabs/mcp/blob/main/README.md

Hope this helps get you started with AWS + MCP!

AWS has introduced several services that support the Model Context Protocol (MCP), which is an open protocol for standardizing how AI models securely access external tools, data sources, and APIs.

Currently, AWS offers MCP servers for:

1. AWS Serverless (AWS Lambda)
2. Amazon Elastic Container Service (Amazon ECS)
3. Amazon Elastic Kubernetes Service (Amazon EKS)
4. Finch

These MCP servers enhance AI-assisted development by providing AI code assistants with real-time, contextual understanding of AWS services. They incorporate AWS operational best practices, Well-Architected principles, and service-specific optimizations.

Additionally, Amazon Q Developer CLI now supports MCP, allowing developers to leverage a wider range of tools and prompts for richer contextual development workflows. This integration enables Amazon Q to provide more customized responses by orchestrating tasks across native and MCP server-based tools.

AWS is also working on integrating MCP with Amazon Bedrock, which would streamline integration between language models and AWS data services. For AWS customers, MCP represents an opportunity to use existing AWS security mechanisms like IAM for consistent access control and to build composable, scalable AI solutions.

To get started with MCP on AWS, you can explore the aws-labs GitHub repository, which contains installation guides, example configurations, and additional specialized servers.
Sources
Enhance AI-assisted development with Amazon ECS, Amazon EKS and AWS Serverless MCP server | AWS News Blog
Announcing new Model Context Protocol (MCP) Servers for AWS Serverless and Containers - AWS
Unlocking the power of Model Context Protocol (MCP) on AWS | AWS Machine Learning Blog
Amazon Q Developer CLI now supports Model Context Protocol (MCP) - AWS

Dear Yan Shum,

I wanted to share some helpful resources regarding Model Context Protocol (MCP) server implementation:

For a detailed walkthrough on using MCP server (EKS) with Amazon Q and VS Code, please refer to this comprehensive guide: https://repost.aws/articles/AR8_eCUcF6QHK8BZrzK4zxHw/managing-eks-clusters-with-amazon-q-and-the-eks-mcp-server-in-your-ide

You can find all AWS-provided MCP servers in our official repository: https://awslabs.github.io/mcp/

These resources should help you implement MCP servers according to your requirements. The EKS example in the article can serve as a template for implementing other MCP servers as well.

Thank you!

Please upvote if you are satisfied with the answer!