Repolode
Submit
awslabs

awslabs/mcp

A collection of open source MCP servers that give AI assistants access to AWS documentation, services, and tooling.

View on GitHubWebsite
Stars
9,797
Forks
1,816
Open issues
241
Last commit
2 days ago

What is it?

This repository is a suite of Model Context Protocol (MCP) servers for AWS. Each server is a small program that exposes a specific set of AWS capabilities, such as documentation search, infrastructure-as-code guidance, or container and serverless management, to MCP-compatible clients like Kiro, Cline, Cursor, Windsurf, VS Code, and Claude Desktop.

The servers cover a broad range of AWS areas, including documentation access, CloudFormation and CDK guidance, EKS and ECS deployment, Lambda and serverless application workflows, Amazon Bedrock knowledge bases, Kendra and Q Business search, and cost and operations tooling. Installation is typically done through uvx, and the servers communicate over stdio. The project notes that Server-Sent Events support was removed in the latest major versions and that Streamable HTTP support is in progress.

The repository is a predecessor to the Agent Toolkit for AWS, and its most useful projects are expected to move there over time. It continues to accept contributions.

Key features

  • 01Provides real-time access to official AWS documentation and API references.
  • 02Offers infrastructure-as-code tools for CloudFormation and CDK.
  • 03Includes servers for managing EKS, ECS, and serverless applications.
  • 04Connects to Bedrock knowledge bases, Kendra, and Amazon Q Business for search.
  • 05Works with MCP clients including Kiro, Cursor, VS Code, and Claude Desktop.

Use cases

  • Letting a coding assistant answer questions about current AWS services and APIs while writing code.
  • Generating and reviewing CDK or CloudFormation templates with AI assistance.
  • Deploying and troubleshooting containerized or serverless applications on AWS from an AI client.
  • Adding enterprise search over AWS-hosted content to an AI assistant.

Quick start

  1. 1

    Pick a server from the Available MCP Servers tables in the README

  2. 2

    Use the install button for your client (Kiro, Cursor or VS Code) or follow the client setup in Installation and Setup

  3. 3

    Restart your MCP client so it loads the new server configuration

Things to know

  • These servers support only the stdio transport, not streamable HTTP or SSE.
  • The README says the most useful projects here will move into Agent Toolkit for AWS, which is the successor to these servers.

Topics

#aws#mcp#model-context-protocol#ai-agents#infrastructure-as-code#serverless#cloud#mcp-client#mcp-clients#mcp-host

FAQ

What is mcp?
A collection of open source MCP servers that give AI assistants access to AWS documentation, services, and tooling. This repository is a suite of Model Context Protocol (MCP) servers for AWS. Each server is a small program that exposes a specific set of AWS capabilities, such as documentation search, infrastructure-as-code guidance, or container and serverless management, to MCP-compatible clients like Kiro, Cline, Cursor, Windsurf, VS Code, and Claude Desktop.
Is mcp free to use? What license does it have?
mcp is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
Is mcp still maintained?
Actively maintained. The last commit to mcp was on Oct 20, 2026.
What can I use mcp for?
Letting a coding assistant answer questions about current AWS services and APIs while writing code. Generating and reviewing CDK or CloudFormation templates with AI assistance. Deploying and troubleshooting containerized or serverless applications on AWS from an AI client. Adding enterprise search over AWS-hosted content to an AI assistant.
How do I install mcp?
Pick a server from the Available MCP Servers tables in the README Use the install button for your client (Kiro, Cursor or VS Code) or follow the client setup in Installation and Setup Restart your MCP client so it loads the new server configuration
Where can I get mcp?
The source code and installation instructions are on GitHub: https://github.com/awslabs/mcp

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects