A new AI-powered integrated development environment (IDE) called Kiro has entered public preview, in a bid to position itself as a tool that helps developers move beyond prototype coding to production-ready systems. The agentic IDE, built on the Code OSS framework, aims to close the gap between idea and deployment by introducing a structured development model based on “specs” and “hooks”.
Also read: AWS Announces Launch of Strands Agents
According to the company, the IDE is designed to address a common pain point: the ease of creating functional applications through prompting is often followed by the harder, less-documented work of preparing that code for production. Developers are left guessing how AI-generated components were built, what assumptions the system made, or whether the output aligns with intended requirements. The AI IDE attempts to formalize that journey by introducing spec-driven development as a workflow layer.
Specs in Kiro serve as living artifacts that represent detailed planning and behavioral intent, says the company in its official blog. From a single prompt, like “Add a review system for products”, Kiro can generate user stories using EARS (Easy Approach to Requirements Syntax), complete with acceptance criteria. These specs guide the AI agent’s implementation and act as a reference point for teams. The AI IDE also creates technical designs based on the approved requirements, including data flow diagrams, interfaces, database schemas, and APIs.
The IDE then breaks down the development effort into a series of interconnected tasks and sub-tasks, each linked back to the spec and including test coverage, mobile and accessibility considerations, and UI states. Developers can implement or review each step individually, with audit trails and code diffs available throughout.
Another central feature of Kiro is its “hooks” system, which are event-driven automations that are triggered by actions like saving or modifying a file. These hooks can, for example, auto-update test files when components are edited, scan for credential leaks before commits, or refresh documentation after endpoint changes. The idea is to enforce team-wide standards and prevent oversights that often arise during fast-moving projects.
Kiro is positioning itself as more than just a prompt-based coding assistant. It incorporates familiar IDE features such as agentic chat, steering rules for AI behavior, Model Context Protocol (MCP) support, and compatibility with existing VS Code settings and Open VSX plugins. But its main pitch is helping developers avoid messy handoffs and missing documentation, particularly when building production-grade systems or collaborating across teams.
While the product is in free preview with usage limits, it supports macOS, Windows, and Linux, and is compatible with several mainstream programming languages. A step-by-step tutorial is also available for developers who want to test the IDE with a complete feature implementation right from initial prompt to deployment.
Kiro joins a growing segment of tools designed to make AI-assisted software development more maintainable and production-ready. Rather than promising instant results, its focus is on better documentation, design clarity, and consistent execution, which are factors that have traditionally separated working prototypes from sustainable products.
