A new player has entered the AI-powered integrated development environment (IDE) space: AWS Kiro, an experimental, agentic IDE designed to streamline the journey from prototype to production. Unlike traditional AI coding assistants that respond to single prompts, Kiro is built with advanced AI models capable of autonomous, goal-driven actions, grounded in context and capable of multi-step reasoning. Instead of generating code snippets one at a time, developers can describe what they want to build, and Kiro actively investigates the codebase, opens relevant files, and modifies them accordingly to fulfill the request.

Kiro facilitates a spec-driven development methodology where intentions are communicated through natural language and diagrams, reducing the need for endless prompt tweaking. This approach allows Kiro to "understand what you're actually trying to build," delivering better results with fewer iterations. These specifications serve as useful artifacts for in-depth feature planning, refactoring, or understanding system behavior.

Under the hood, Kiro operates on a few core principles:

• Agentic Reasoning Loop: It uses a structured loop of planning, reasoning, taking actions, and evaluating results to handle multi-step tasks.
• Context Awareness: It integrates tightly with the local environment using the MCP protocol and local LSP (Language Server Protocol) tools.
• Security and Privacy: All actions are transparent, and code execution is done locally, ensuring data and code are not sent to the cloud unless explicitly allowed. Users can also choose between "Autopilot" mode, where the agent can freely make code changes, and "Supervised" mode, which requires approval before any change.

Kiro's key features include Model Context Protocol (MCP) support for connecting specialized tools, steering rules to guide AI behavior across projects, and agentic chat for ad-hoc coding tasks with file, URL, and Doc's context providers. It is built on Code OSS, allowing users to keep their VS Code settings and Open VSX-compatible plugins. Kiro supports popular languages like Python and JavaScript, with support for additional languages actively being developed. It also helps maintain up-to-date dependencies automatically.

Kiro isn't just about AI assistance; it's about intelligent orchestration of the entire project development lifecycle. This spec-driven approach ensures that every line of code serves a clear purpose, every architectural decision is documented, and every task builds logically toward the goals.

While Kiro integrates with AWS services, it operates agnostically with any technology stack and cloud provider. Users can get started with Kiro using Google Sign-in, GitHub, Builder ID, or AWS SSO authentication; an AWS account is not mandatory.

Kiro is currently in public preview, with free access (with some limits). After the preview period, there will be a free tier with 50 agentic interactions per month, a Pro account for $19.00 per user/month with 1,000 interactions, and a Pro+ account for $39.00 with 3,000 interactions. Additional interactions will cost $0.04 each.