AWS Kiro, a new AI IDE, has been introduced to streamline software development from concept to production, focusing on “spec-driven development.” Announced by Nikhil Swaminathan and Deepak Singh, Kiro aims to bridge the gap between rapid prototyping and robust production systems.
The core of Kiro’s functionality lies in its “specs” and “hooks.” Kiro specs are artifacts that guide AI agents by unpacking requirements from a single prompt, generating detailed user stories with EARS (Easy Approach to Requirements Syntax) acceptance criteria. It then creates comprehensive design documents, including data flow diagrams, TypeScript interfaces, database schemas, and API endpoints, based on approved requirements and codebase analysis.
Kiro further generates and sequences tasks and sub-tasks, linking them back to requirements and including details such as unit tests, integration tests, and accessibility considerations. Kiro hooks are event-driven automations that act as background collaborators, triggering agents to perform tasks like updating test files, refreshing documentation, or scanning for security vulnerabilities upon file saves or creation.
This ensures consistency and enforces coding standards across development teams. Beyond these features, Kiro includes Model Context Protocol (MCP) support, steering rules for AI behavior, and agentic chat. Built on Code OSS, it allows developers to retain existing VS Code settings and Open VSX compatible plugins.
Kiro is currently available for free during its preview phase, supporting Mac, Windows, and Linux, and multiple programming languages. Users can download Kiro and connect via Discord or social media platforms using #builtwithkiro.
