Overview

Kiro is an AI-powered, agentic IDE (integrated development environment) that transforms natural language prompts into structured software specifications and working code. Agent Toolkit for AWS is a production-ready suite of tools and guidance that gives AI coding agents the tools, knowledge, and guardrails they need to work effectively with AWS services, which helps to improve success rates and reduce wasted effort.

This article shows how to setup Kiro IDE with Agent Toolkit for AWS , and use them to manage your AWS environment with natural language prompts. It also suggests how you can extend your Kiro AWS capabilities with Open Source MCP Servers for AWS and Kiro Powers

Similar articles on Agent Toolkit for AWS are available for Claude Code and OpenAI Codex

What can I do with Agent Toolkit for AWS?

Agent Toolkit for AWS includes

• AWS MCP Server — A managed server that gives agents access to AWS through the Model Context Protocol (MCP)
• Agent skills — Curated packages of instructions, code scripts, and reference materials that help agents complete specific AWS tasks
• Plugins — Single-install packages for Claude Code and Codex that bundle the AWS MCP Server configuration and a curated set of agent skills
• Rules files — Project-level configuration files that set guardrails and preferences for how agents work in your project

You can use Agent Toolkit for AWS to do the following

• Get real-time AWS knowledge – Search and retrieve up-to-date AWS documentation, API references, best practices, and regional availability information to inform your AI assistant's responses and decisions.
• Make authenticated AWS API calls – Execute most of the 15,000+ AWS APIs with SigV4 through your existing IAM roles and policies, with automatic syntax validation and error handling.
• Troubleshoot AWS issues – Analyze CloudWatch logs and CloudTrail events, investigate permission problems, debug application failures, and diagnose performance issues using guided workflows and access to comprehensive AWS knowledge sources.
• Provision and configure infrastructure – Create and configure AWS resources like VPCs, databases, compute instances, and storage with automated workflows that implement security best practices and proper resource tagging.

AI agent setup prompt

Besides following the guidance in this article, you can also setup Agent Toolkit with the following prompt

Set up Agent Toolkit for AWS by following instructions:
https://raw.githubusercontent.com/aws/agent-toolkit-for-aws/refs/heads/main/setup-instructions/setup.md

Requirements

Operating System (OS)

Kiro is available for macOS, Windows and Linux

• macOS: Kiro runs on both Intel and Apple silicon with the latest security updates
• Windows: Kiro runs on 64-bit Windows. ARM is not currently supported
• Linux (x86_64): Kiro requires glibc 2.39 or higher, e.g. Ubuntu 24+, Debian 13+, Fedora 40+, Arch Linux, and Linux Mint 22+

Software

Besides Kiro, the following software are required

• AWS Command Line Interface (AWS CLI)
• UV Python package and project manager

Install Software

Install AWS CLI

Installation files

To install, you can download latest binaries/installer

• Windows : https://awscli.amazonaws.com/AWSCLIV2.msi
• Mac OS: https://awscli.amazonaws.com/AWSCLIV2.pkg
• Linux: https://awscli.amazonaws.com/awscli-exe-linux-x86_64.zip

Do refer to AWS CLI documentation for installation guidance and other installation options.

To verify

aws --version

Output should be similar to below

aws-cli/2.35.20 Python/3.14.6 Windows/11 exe/AMD64

Configure AWS CLI

You need to configure AWS credentials on your local machine. The server uses these credentials to authenticate your requests.

Option 1: AWS Management Console credentials

If you have username and password that allows you to sign in to the console.

aws login

Option 2: AWS SSO (IAM Identity Center)

If you login using IAM Identity Center

aws configure sso

Follow the prompts to set up your SSO configuration.

Option 3: IAM User

If you have a IAM user

aws configure

Enter your Access Key ID, Secret Access Key, default region, and output format.

Option 4: EC2 Instance

On EC2 instances with no AWS profiles configured, AWS CLI automatically and securely retrieves the credentials from the instance metadata without additional configuration.

If you wish to configure a profile, for example EC2, that uses EC2 instance metadata

aws configure set --profile EC2 credential_source Ec2InstanceMetadata
aws configure set --profile EC2 region ap-southeast-1

Do replace ap-southeast-1 with the correct Region Code

AWS profile

Do note the profile name. If you do not specify one, the profile name is usually default. You can retrieve profile name from ~/.aws/config file or with the following command

aws configure list-profiles

Output could be similar to below

default

Verify

aws sts get-caller-identity

You should get a output with details about your IAM user or role

{
    "UserId": "AROAXD2DB4IT6MCPEU35:awsadmin",
    "Account": "123456789012",
    "Arn": "arn:aws:sts::123456789012:assumed-role/AWSReservedSSO_AdministratorAccess_4eaed2acd485a10b/awsadmin"
}

Install UV

Windows

From PowerShell

powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

Linux / MacOS

curl -LsSf https://astral.sh/uv/install.sh | sh

Verify

uvx --version

You should get output that list UVX version, e.g.

uvx 0.11.7 (9d177269e 2026-04-15 x86_64-pc-windows-msvc)

Do refer to UV documentation for details including other installation options

Install Kiro

Installation

• Go to kiro.dev or Kiro downloads, and download IDE installer. CLI is optional.
• Open the downloaded file and follow the installation instructions

Sign In

To use Kiro, you will need to Sign In.

Kiro supports several authentication options including GitHub, Google, AWS Builder ID, AWS IAM Identity Center and external identity provider such as Microsoft Entra ID or Okta

Once you have sign-in, Open a Project and select a folder

Select Yes, I trust the authors if prompted

Demo video

Install AWS MCP Server

• Open the Kiro panel
• In the MCP servers section, select the Open MCP Config icon

Alternatively, you can use AWS CLI to install default AWS skills, and configure the AWS MCP Server. Refer to Skills section below for more information

In the mcp.json editor window, paste the following text

{
  "mcpServers": {
    "aws-mcp": {
      "command": "uvx",
      "timeout": 100000,
      "transport": "stdio",
      "args": [
        "mcp-proxy-for-aws@latest",
        "https://aws-mcp.us-east-1.api.aws/mcp",
        "--metadata",
        "AWS_REGION=us-west-2"
      ],
      "autoApprove": [
        "aws___retrieve_skill",
        "aws___search_documentation",
        "aws___search_documentation",
        "aws___read_documentation",
        "aws___recommend",
        "aws___list_regions",
        "aws___get_regional_availability",
        "aws___call_aws",
        "aws___suggest_aws_commands",
        "aws___run_script",
        "aws___get_presigned_url",
        "aws___get_tasks"
      ]
    }
  }
}

Do replace us-west-2 with the correct AWS Region Code. Entries in autoApprove section are optional

If you want to use a different AWS profile or multiple profiles, add AWS_MCP_PROXY_PROFILES environment parameter in mcp.json

{
  "mcpServers": {
    "aws-mcp": {
      "command": "uvx",
      "timeout": 100000,
      "transport": "stdio",
      "args": [
        "mcp-proxy-for-aws@latest",
        "https://aws-mcp.us-east-1.api.aws/mcp",
        "--metadata",
        "AWS_REGION=us-west-2"
      ],
      "env":{
        "AWS_MCP_PROXY_PROFILES": "default dev prod"
      },
      "autoApprove": [
        "aws___retrieve_skill",
        "aws___search_documentation",
        "aws___search_documentation",
        "aws___read_documentation",
        "aws___recommend",
        "aws___list_regions",
        "aws___get_regional_availability",
        "aws___call_aws",
        "aws___suggest_aws_commands",
        "aws___run_script",
        "aws___get_presigned_url",
        "aws___get_tasks"
      ]
    }
  }
}

Refer to MCP Proxy for AWS Python package page for more details

Save the file and close Kiro

Verify MCP Server

Launch Kiro, and wait for aws-mcp status to change to Connected

If you do not get Connected status, re-verify your AWS CLI profile credentials

Install Rules file

Agent Toolkit for AWS includes a recommended steering file that tells your agent how to work with AWS — for example, to use the AWS MCP Server for API calls, to search for available skills before starting a task, and to prefer infrastructure-as-code over direct CLI command

Create

To install in Kiro, open Kiro panel, under Agent Steering & Skills section, click the Add (+) icon

Select Global agent steering

Enter aws-agent-rules and Press `Enter'

Copy and paste raw contents of aws-agent-rules.md from GitHub

The recommended rules file is a starting point. You can customize it for your project — for example, specifying which AWS Region to use, which VPC to deploy into, or which naming conventions to follow for resources.

Save the file

Verify

You can verify by asking Kiro AWS related questions and look for Including Steering Documents message

Install Agent Skills

Agent skills are curated packages of instructions, code scripts, and reference materials that help AI coding agents complete specific AWS task. Skills bridge the gap between what AI models know from their training data and what is needed to work effectively with AWS — especially for newer services, complex multi-service workflows, and tasks where best practices matter.

Option 1: Using AWS CLI v2

AWS CLI version 2.35.0 or later can be used to install Agent skills and/or AWS MCP server from command line

aws configure agent-toolkit --region us-east-1

Output should be similar to follow

Detecting installed AI coding agents...
  ✗ Claude Code — ~/.claude/skills (not found)
  ✗ Cline — ~/.cline/skills (not found)
  ✗ Codex — ~/.agents/skills (not found)
  ✗ Cursor — ~/.cursor/skills (not found)
  ✗ Gemini CLI — ~/.agents/skills (not found)
  ✓ Kiro — ~/.kiro/skills
  ✗ OpenClaw — ~/.openclaw/skills (not found)
  ✗ OpenCode — ~/.agents/skills (not found)
  ✗ Pi — ~/.pi/agent/skills (not found)
  ✗ Windsurf — ~/.agents/skills (not found)
 
Select agents to configure (space to toggle, up/down arrows to navigate, enter to confirm):
  [x] Kiro — ~/.kiro/skills

Follow on screen instructions to install core skills, and optionally install AWS MCP server

Updating, adding and removing skills with AWS CLI

After which, you can use aws agent-toolkit command to install, update, remove, list, and search for individual AWS skills

To list installed skills

aws agent-toolkit --region us-east-1 list-installed-skills

To list available skills

aws agent-toolkit --region us-east-1 list-available-skills

To install a skill, e.g. cloudfront

aws agent-toolkit --region us-east-1 add-skill --skill-name cloudfront

To update a skill, e.g. amazon-bedrock

 aws agent-toolkit --region us-east-1 update-skill --skill-name amazon-bedrock

Refer to AWS CLI integration guide for more details, and Agent Toolkit Github for listing of specialized skills

Option 2: Using Node

Software requirements

You need

• Node.js
• Git client

Install Node.JS

To install, refer to Node.js documentation

To verify

node --version

Output should be similar to below

v26.0.0

Install Git client

To install, refer to Git documentation

To verify

git --version

Output should be similar to below

git version 2.54.0.windows.1

Windows

For Windows, enable long paths to allows Git to bypass the standard Windows maximum path length limitation of 260 characters

git config --global core.longpaths true

From Kiro, open a terminal in your project folder and run the following command

npx skills add aws/agent-toolkit-for-aws/skills

Follow screen instructions to install

Verify

To verify what skills have been installed, ask Kiro

What AWS skills do you have available?

Skills that are installed on global level are listed in Kiro panel under Agent Steering & Skills

Refer to documentation for more information

Using Kiro with Agent Toolkit for AWS

Using Kiro with AWS MCP server

Use Kiro chat window to ask questions about AWS and your AWS environment. Below are some examples

Search AWS Documentation

For example, you can ask about Bedrock AgentCore. Kiro will use AWS MCP server to search AWS documentation as indicated by the Called MCP tool aws-mcp message

Retrieve information about your AWS environment

To retrieve Python versions that your Lambda functions are running on

Update AWS configuration

To update EC2 instance configuration

Optional: Open Source MCP Servers for AWS

The AWS MCP Server consolidates capabilities from existing MCP servers (AWS Knowledge MCP and AWS API MCP) into a single, unified interface. Other specialized MCP servers are available to help you get most out of your AWS environment.

Install MCP servers for AWS

Go to Open Source MCP Servers for AWS to view list of available MCP servers

Browse and select the specific MCP server you are interested in. Most of them have a Add to Kiro icon that you can use to install the MCP server

Do refer to your MCP server documentation page for additional installation instructions (if any). If most cases, you will need to configure AWS_PROFILE and AWS_REGION values

Optional: Kiro Powers

Kiro powers give your AI agent instant access to specialized knowledge for any technology. Powers package your tools, workflows, and best practices into a format that Kiro can activate on-demand. When you mention relevant keywords, Kiro loads the power's context and tools automatically.

Power Listing

You can find a list of powers at https://kiro.dev/powers/

Click on + Add to Kiro to install your desired Kiro power

Verify

In the Kiro output screenshot power, HealthOmics power is activated to migrate a genomics workflow to AWS HealthOmics

To install and learn more about HealthOmics power, refer to Install AWS HealthOmics Kiro Power to create, migrate, run, debug and optimize genomics workflows

Additional Information

Blog post The AWS MCP Server is now generally available provides an overview of AWS MCP Server.

Refer to Kiro IDE documentation, Agent Toolkit for AWS User Guide and Open Source MCP Servers for AWS for more information

While it provides a free tier, Kiro users with paid enterprise subscription plans have access to more powerful AI models. To charge your Kiro subscription plan to your AWS account, refer to Enable Kiro enterprise subscription with IAM Identity Center in your AWS account