KIRO - The intelligent AWS assistant that's transforming the way we work with code.
In the world of AI programming tools, a new player has emerged that is changing how we write, document, and maintain code. KIRO, developed by Amazon Web Services, is an intelligent development environment that combines the familiar convenience of a code editor with the powerful capabilities of artificial intelligence—and it does so in a completely new way.
What is KIRO?
KIRO is an intelligent integrated development environment (IDE) built on Visual Studio Code (Code OSS) that uses advanced AI models such as Claude Sonnet 3.7 and 4.0 from Anthropic and Amazon Bedrock (Nova). It supports developers at every stage of the application lifecycle from planning and implementation to monitoring, documentation, and testing.
This is not just another AI code assistant. It’s a full development environment introducing concepts like spec-driven development, an advanced hook system, configurable project guidelines (known as steering), and contextual conversations with AI agents that understand not only your code, but the logic behind your project.
What can KIRO do?
KIRO is a fully capable agent coding partner. It supports numerous programming languages including Python, Java, JavaScript, TypeScript, C#, Go, Rust, PHP, Ruby, Kotlin, C, C++, shell scripting, SQL, Scala, JSON, YAML, and HCL. Interaction with the AI agent is currently available in English. Here’s what KIRO offers:
Spec-driven development
KIRO introduces a structured approach based on three key files: requirements.md, design.md, and tasks.md. Based on these, it automatically generates code, tests, user stories, architecture diagrams, and database schemas. All elements are contextually linked and can be updated with a single click.
Agent Hooks
In KIRO, you can define custom hooks that execute tasks upon saving a file—such as generating unit tests, updating documentation, performing security scans, or completing the README. This enables seamless automation of repetitive tasks.
Autopilot and supervised mode
Built-in AI agents can make code changes automatically in autopilot mode or operate in supervised mode, where every change requires manual approval. This setup ensures full control over what enters the repository.
Steering – custom project rules
You can create so-called steering files—Markdown documents defining standards for code style, architecture, naming, technologies used, or dependencies. This is an excellent solution for teams seeking consistency regardless of the number of developers involved.
Model Context Protocol (MCP)
KIRO can pull context from external sources such as AWS documentation, GitHub, Jira, or Asana. This eliminates the need to switch between applications—everything is accessible directly in the IDE with full project context.
How much does KIRO cost?
Currently, KIRO is available for free in its public preview version, allowing users to explore all features before official subscription plans are introduced.
In the future, Amazon Web Services plans to implement a pricing model based on tiered packages. On the tool’s website, you can already see three pricing tiers, with the PRO version starting at $19/month per user.
Users of Amazon Q Developer and Amazon Q Developer Pro will retain access to KIRO at no extra cost. Pro users can log in to KIRO using their AWS IAM Identity Center credentials.
Summary
KIRO is a tool that has the potential to significantly transform how programming is done. While it’s still in testing and available for free, it’s worth trying out to see what it can do in real development scenarios.
Follow our profile on LinkedIn to stay up to date with the latest tech trends and solutions that truly support IT teams in their work.
