Kiro is an AI-integrated development environment created by Amazon Web Services. It introduces a spec-driven development workflow in which a developer's prompt is first expanded into a requirements document, then a design document, and finally a set of implementation tasks that AI agents execute. This structured approach aims to reduce ambiguity and keep AI-generated code aligned with the original intent throughout the development lifecycle. Kiro includes an agentic coding assistant, automated hooks that run on file-save events to enforce linting, testing, and documentation, and a steering-rules system that lets teams encode project-wide conventions the AI must follow. It is built on the VS Code open-source foundation, supports standard VS Code extensions, and connects to Model Context Protocol (MCP) servers for external tool access. Kiro was made available as a free preview in July 2025 and targets individual developers as well as development teams looking to accelerate software delivery with AI assistance.
Target audience and deployment
- Solo / Freelancer
- Startup
- SMB
- Mid-market
- Enterprise
- Cloud
Key features
Use cases
- Generate production-ready code from natural language prompts
- Enforce project-wide coding conventions with steering rules
- Automate repetitive tasks on file save with hooks
- Collaborate with AI agents on multi-step software design
- Extend IDE capabilities via MCP servers
Best for
- Individual developers who need to accelerate feature delivery with AI-assisted coding
- Development teams who need to enforce consistent coding standards across AI-generated code
- Software engineers who need to translate product requirements into implementation tasks automatically
Integrations
AI models included
Amazon Bedrock, Claude
Other
Model Context Protocol (MCP)