The autonomous agent that extends your flow

An agent is an AI system that can plan and execute multi-step tasks with some degree of autonomy. Unlike simple chatbots that respond to individual prompts, agents can break down complex tasks, make decisions about how to proceed, and take actions to accomplish goals.

Frontier agents are a new, more sophisticated class of AI agents offered by AWS with defining characteristics. They’re autonomous: direct them towards a goal, and they figure out how to achieve it. They’re massively scalable, able to perform multiple concurrent tasks and distribute work across agents. And, they work independently, operating for hours or days without intervention.

Kiro autonomous agent works independently on tasks without requiring constant human guidance. In the context of development, agentic development means the AI can take a high-level task description, figure out the implementation plan, write code across multiple repositories, run tests, and create pull requests—all while you work on other things. The agent operates asynchronously in the background rather than requiring you to stay in an active session.

We’re starting to roll out the preview to Kiro Pro, Pro+, and Power users now. Access will be gradually expanded over the coming weeks. Check kiro.dev to see if the feature is available for your account.

If you’re a Kiro Pro, Pro+, or Power user, you’ll see the new autonomous agent tab on Kiro.dev once it’s rolled out to your account. We’re starting the rollout now and will be gradually expanding access during the preview period.

The autonomous agent isn’t enabled until you set it up by connecting your GitHub account and assigning it tasks. If you prefer not to use it, simply don’t complete the setup. It won’t affect your existing Kiro IDE or CLI workflows.

There’s no cost during the preview period for Kiro Pro, Pro+, and Power users. Usage is subject to weekly limits that reset each week. We’ll communicate pricing details before the preview period ends.

Kiro autonomous agent creates pull requests for your review, and it never merges changes automatically. We recommend protecting your main and other branches where you don’t want the agent pushing to directly to these branches as it works on coding tasks.

Kiro CLI’s custom agents are configuration files that customize Kiro’s behavior for specific workflows. They define which tools the agent can access, what permissions it has, and what context to include—allowing you to pre-approve tools, reduce interruptions, and optimize for specific tasks during your active development sessions. Kiro autonomous agent, on the other hand, works asynchronously in the background on complex, multi-step development tasks. It runs in isolated sandbox environments, creates pull requests, learns from code reviews, and maintains context across multiple repositories and sessions. CLI custom agents are for configuring interactive workflows on your local machine, while autonomous agents handle independent, long-running development work.

Kiro IDE and Kiro autonomous agent complement each other. Kiro IDE is excellent for active collaboration with agents on your local machine—pair programming, getting suggestions, and iterating on code in real-time. Kiro autonomous agent handles independent, asynchronous work in the background. You assign tasks to Kiro autonomous agent from kiro.dev or GitHub, not from within the IDE. You might use Kiro IDE for your primary development work while allowing Kiro autonomous agent to spend more time on complex reasoning tasks, taking into consideration the context learned from the repositories and pull requests. The autonomous agent learns from your code review feedback and applies those patterns, which can inform how you work in the IDE as well.

Kiro CLI and Kiro autonomous agent serve different purposes. Kiro CLI is great for working directly with agents on your local machine during active development sessions. Kiro autonomous agent works asynchronously in the background on tasks you assign from kiro.dev or GitHub. You might use Kiro CLI for interactive development and pair programming, while using Kiro autonomous agent for tasks like updating dependencies across 15 microservices or implementing features from your backlog while you focus on other work.

During preview, Kiro autonomous agent for teams integrates with Jira, Confluence, GitLab, GitHub, Teams, and Slack. Teams request access and are onboarded in phases as we expand coverage.