Kiro is an agentic engineering platform built by AWS that pushes past ad-hoc "vibe coding" prompts toward spec-driven development, turning a prompt into structured requirements, a design, and a sequenced task list before agents start implementing.
Who builds it
Kiro is built and operated by a small, opinionated team within AWS, on AWS infrastructure. The name is meant to evoke "a hardworking person who never gives up," reflecting the team's stated goal of iterating quickly on an AI-powered development experience rather than shipping a single, static assistant.
Core features
- Spec-driven development: converts an initial prompt into requirements, an architectural design, and a sequenced task list before any code is written, instead of jumping straight to generation.
- Correctness validation: uses property-based testing to catch edge cases that conventional unit tests can miss.
- Parallel agents: multiple agents can work on a codebase at once and carry learning across sessions rather than starting from zero each time.
- Multiple surfaces: available as an IDE, a CLI, a web browser experience, and a mobile app, so work can be reviewed or steered from different contexts.
- Tool integrations: connects to Figma, Terraform, GitHub, GitLab, and CI/CD systems as part of a spec-to-implementation workflow.
- Broad language support: Python, JavaScript, TypeScript, Java, Rust, Go, and other common languages.
- Enterprise controls: IAM/SSO authentication, usage dashboards, cost controls, and governance features built on AWS infrastructure.
Pricing
Kiro uses a credit-based model with no daily or weekly rate limits, letting you choose among multiple AI models (including Claude Opus, Sonnet, and Haiku, among others) at different credit costs:
- Kiro Free: $0/month — 50 credits, with access to open-weight models and Claude Sonnet 4.5.
- Kiro Pro: $20/month — 1,000 credits, premium model access, add-on credits at $0.04 each.
- Kiro Pro+: $40/month — 2,000 credits, same premium model access and add-on rate.
- Kiro Pro Max: $100/month — 5,000 credits.
- Kiro Power: $200/month — 10,000 credits.
- Team plans mirror the individual pricing tiers per user, adding consolidated billing, usage analytics, SSO via AWS IAM Identity Center, and organizational dashboards.
- Unused monthly credits expire and don't roll over; new users signing up via social login or AWS Builder ID get a $20 signup bonus.
Who it's for
Teams and individual developers who want AI agents to work from an agreed specification rather than a loose prompt — useful for larger codebases where an unreviewed "vibe coded" change is risky, and for organizations already standardized on AWS that want IAM/SSO-based governance over who can run agents and how much they can spend. The free tier's 50 credits are enough to trial the spec-driven workflow before committing to a paid plan.