AWS open-sources Kiro Crew: autonomous AI agents for dev teams
AWS released Kiro Crew, an open-source orchestration platform for running autonomous, multi-agent engineering workflows across repositories, developer tools, and multiple work sessions. The goal is persistent background work, not just assisted code generation.
What Kiro Crew Does
Rather than waiting for a developer to prompt it, Kiro Crew coordinates multiple AI agents simultaneously, schedules recurring tasks, preserves project context between sessions, and integrates with the tools already in your dev stack. The practical use cases according to AWS: incident investigation, PR monitoring, ticket triage, dependency upgrades, framework migrations, and flaky test cleanup.
AWS Distinguished Developer Advocate Darko Mesaros described it as “an application layer that turns AI coding agents into always-working, self-learning, autonomous teammates.”
Origin and Adoption
The project started inside Amazon as an internal tool called MeshClaw. More than 39,000 Amazon builders adopted it in under six months before AWS open-sourced it.
It ships with persistent memory, multi-agent orchestration, approval workflows, scheduling, sandboxing, signed audit logs, and a web and desktop dashboard. It can run entirely inside customer environments: laptops, containers, or VMs. No AWS account required, no AWS-managed control plane.
Reference Apps Launching With It
- DevFleets: worktree management
- Issue Radar: issue and pull-request triage
- Task Runner: long-running engineering task execution
These are purpose-built interfaces layered on top of Kiro Crew’s orchestration engine, not standalone products.
The Catch
Kiro Crew is built around open standards (ACP and MCP), but at launch it runs on the proprietary Kiro CLI, which is metered by credits. Enterprises running Claude Code, Codex, or Devin will need to build and validate their own connectors before using Kiro Crew as their orchestration layer. Teams already using Kiro can reuse existing .kiro configurations, including steering files, skills, and custom agents, without additional setup.
The platform itself is free. You pay only for the AI coding agents and tools you connect to it. AWS will govern the project through a public steering committee operating under an open governance model.
I’m founder of BizStack at Cagri Sarigoz LLC and a passionate advocate for entrepreneurs.
With over 14 years in tech, marketing, and AI, including my role as Head of SEO at CitizenShipper and co-founder of TaleBot at Intale AI, I’m dedicated to sharing genuine, useful product insights and tips.
At BizStack, I aim to cut through the digital noise to provide clear, actionable advice.
And more than all else, I’m a father to a (always) little girl and a husband.
Contact me at [email protected] for assistance.
