Free
Basic use
Spec-driven agentic IDE by AWS — from prompt to spec to production.
Kiro is AWS's spec-driven agentic IDE — the successor to Amazon Q Developer. Instead of jumping straight to code, Kiro transforms natural language prompts into structured specs (user stories, acceptance criteria, data flow diagrams) before writing a single line. Built on Amazon Bedrock and powered by Claude Sonnet 4.5, it adds Hooks (automated agent triggers on file events) and Steering files (persistent project context) for sustained agentic workflows.
Basic use
Advanced interactions, credit-based
Customer-managed encryption keys, admin controls
Kiro is AWS's new agentic IDE, launched at AWS re:Invent 2025 as the direct successor to Amazon Q Developer for IDE workflows. Its defining concept is spec-driven development: rather than accepting a vague natural language prompt and immediately generating code, Kiro first produces a structured specification — user stories with acceptance criteria, a design document, data flow diagrams, and a sequenced implementation plan. The developer reviews and refines the spec before any code is written, ensuring the AI's understanding matches the intent. This positions Kiro closer to a software architect than an autocomplete engine.
Two features distinguish Kiro from other agentic IDEs. Hooks are automated triggers that execute predefined agent actions when specific file-system events occur — saving a file, creating a new file, or deleting one — enabling fully automated workflows without manual re-prompting (e.g., auto-run tests when /src files change, auto-update docs when API routes are modified). Steering files are persistent markdown files checked into the project that give Kiro ongoing knowledge of the codebase's conventions, libraries, and standards, so developers don't have to re-explain project context in every session.
Kiro is built on Amazon Bedrock and powered primarily by Anthropic Claude Sonnet 4.5. It includes all capabilities from Amazon Q Developer — agentic coding, inline chat, terminal integration, MCP server support, security scanning, and unit test generation — plus the new spec/hook/steering layer. Existing Q Developer subscribers are recommended to migrate to Kiro ahead of the Q Developer IDE EOS on April 30, 2027. Pricing is credit-based: Free tier for basic use, Pro at $19/month for advanced interactions, and Enterprise with customer-managed encryption keys and admin controls.
• Spec-driven development: prompt → spec (user stories, acceptance criteria, diagrams) → code — structured engineering workflow
• Hooks: automated agent triggers on file-system events (save, create, delete) — no manual re-prompting for routine tasks
• Steering files: persistent markdown project context — conventions, libraries, and standards always available to the agent
• Agentic chat: natural language questions, explanations, feature generation, debugging across the codebase
• Amazon Bedrock + Claude Sonnet 4.5: AWS-native model hosting with enterprise data residency
• MCP support: extend the agent with external tools and APIs
• Terminal integration: same agentic capabilities in the CLI
• Security scanning: detects secrets, CVEs, IaC misconfigurations (inherited from Q Developer)
• Unit test generation: auto-generates test suites
• Q Developer migration path: official successor for Q Developer IDE users (EOS April 30, 2027)
• Pricing: Free · Pro $19/mo · Enterprise custom
• Platforms: macOS, Linux, Windows (CLI), VS Code-compatible
Rewind Conversations (/rewind to any earlier prompt and branch off), Model Reasoning Effort Control (/effort with 5 levels), Unified Settings Menu, 88% faster workspace initialization.
OAuth Client ID for MCP servers (connect to Slack, GitHub, Figma without proxy), KIRO_HOME environment variable, configurable V2 TUI keybindings.
Launched Kiro Web at app.kiro.dev — cloud-hosted agentic development with autonomous mode, multi-repo PR coordination, GitHub-native workflow.
Parallel Task Execution (independent tasks run concurrently, up to 4x faster), Quick Plan mode, Analyze Requirements for automated reasoning to catch inconsistencies.
Adaptive Thinking Across Multi-Turn Conversations (model reasoning preserved across turns), fix for silent failure in subagent tool dispatch.
A developer prompts Kiro: 'Add a rate-limited email verification flow to the auth service'. Kiro generates a spec with user stories, acceptance criteria, and a sequence diagram before touching any code. After the developer reviews and approves the spec, Kiro implements the feature across multiple files, writes unit tests, and updates the docs. A Hook automatically runs the test suite every time a file in /auth is saved. AI code review platform for the AI era. Automated code reviews, security scanning, and team analytics across GitHub, GitLab, VS Code, and JetBrains. Used by 300,000+ developers.
AI-powered PR description generator and code review assistant. Automatically writes pull request descriptions, sends stakeholder notifications, creates changelogs, and provides inline code refactoring.
Multi-agent AI coding platform with 12+ agents and 24+ models, featuring Chairman LLM for parallel multi-agent evaluation and end-to-end encrypted inference. Ships across six surfaces: CLI, IDE, Cloud, API, Mobile, and Builder.