Using Kiro CLI on WSL: Setup, Tips & Real-World Examples

I've been using Kiro CLI as my primary AI-powered development assistant on Windows Subsystem for Linux (WSL), and it has fundamentally changed how I work. This post covers how to set it up, what it can do, and real examples of what I've built with it.

What is Kiro?

Kiro is an AI coding assistant that runs directly in your terminal. Unlike browser-based AI tools, Kiro operates within your actual development environment — it can read your files, run commands, interact with AWS, and make changes to your codebase in real-time.

Key capabilities:

• File system access — reads, creates, and edits files directly
• Terminal execution — runs shell commands and scripts
• AWS integration — deploys infrastructure, queries services
• Code intelligence — understands project structure, finds symbols, navigates code
• Multi-step tasks — handles complex workflows autonomously

Setting Up Kiro on WSL

Prerequisites

You need WSL2 with Ubuntu (22.04 or later) running on Windows 10/11:

# Check your WSL version
wsl --version

# My setup
# Ubuntu 22.04.5 LTS
# Kernel: 6.6.87.2-microsoft-standard-WSL2

Installation

# Download and install Kiro CLI
curl -fsSL https://kiro.dev/install.sh | sh

# Verify installation
kiro --version
# 0.6.0

# Start a chat session
kiro chat

AWS Configuration

If you're working with AWS (and you should be), make sure your credentials are configured in WSL:

# Your ~/.aws/config should have profiles
[profile my-dev]
region = eu-central-1
role_arn = arn:aws:iam::ACCOUNT_ID:role/MyRole
source_profile = default

Recommended WSL Setup

# Tools that work great alongside Kiro
sudo apt update
sudo apt install -y python3-pip nodejs npm

# Node.js (for CDK projects)
nvm install 22

# AWS CDK
npm install -g aws-cdk

# Python tools
pip3 install boto3

What Can You Achieve?

Here are real examples of tasks I've accomplished with Kiro in single sessions:

Tips for Effective Use

1. Be Direct with Requests

Kiro works best when you're clear about what you want. Instead of "can you help me with my website?", say "redesign my website as a single-page portfolio, look at example.com for reference."

2. Iterate Quickly

Don't try to get everything perfect in one prompt. Deploy first, then iterate:

# First pass
"build the platform and deploy it"

# Then refine
"remove the contact section"
"make github a button also"
"fix 15+ years not 20+"

3. Let It Handle MFA/Auth

For AWS deployments with MFA, give Kiro your token code when asked. It will handle the assume-role flow and pass credentials to subsequent commands.

4. Use It for Cleanup Too

Kiro is excellent at tearing down infrastructure. It knows the dependency order, empties S3 buckets before deleting stacks, and handles failed CloudFormation states.

5. Trust the Multi-Step Process

Complex tasks often require 10-20 tool calls. Kiro will synthesize CDK, find errors, fix them, redeploy, check logs, and iterate — all without you needing to intervene.

Limitations to Be Aware Of

• No interactive prompts — Can't handle tools that require stdin input (like password prompts). Use non-interactive alternatives.
• Session tokens expire — For long-running tasks, AWS STS tokens may expire. Keep a fresh MFA code handy.
• Large outputs — Very large command outputs may be truncated. Pipe to head or tail when needed.
• Network-dependent — Requires internet connectivity for both the AI service and your AWS/GitHub operations.

My Workflow

Here's how I typically work with Kiro on WSL:

1. Open Windows Terminal → WSL Ubuntu
2. Navigate to project → cd ~/my-project
3. Start Kiro → kiro chat
4. Describe the task — architecture, code, deployment, whatever
5. Review and iterate — approve deployments, provide MFA codes, adjust requirements
6. Done — infrastructure deployed, code committed, docs written

The entire flow stays in the terminal. No context-switching between browser tabs, consoles, or IDEs. Everything happens in one place.

Conclusion

Kiro on WSL gives you an AI pair programmer that actually lives in your environment. It's not generating code snippets for you to copy-paste — it's building, deploying, and iterating alongside you. For cloud engineers and solutions architects, this is a significant productivity multiplier.

The combination of WSL's Linux environment, AWS CLI access, and Kiro's ability to chain complex operations makes it possible to go from idea to deployed infrastructure in minutes rather than hours.