AWS’s Kiro Crew aims to turn AI coding agents into autonomous engineering teams

Published: August 4, 2026 — InfoWorld (Anirban Ghoshal)

AWS released Kiro Crew, an open-source orchestration platform designed to move enterprises beyond interactive AI coding assistants toward long-running, autonomous engineering workflows spanning repositories, developer tools, and multiple sessions.

Capabilities

• Coordinates multiple AI agents; schedules recurring work; preserves project context across sessions
• Integrates with developer tools for incident investigation, PR monitoring, ticket triage, and unattended engineering tasks
• Ships with persistent memory, multi-agent orchestration, approval workflows, scheduling, sandboxing, signed audit logs, web/desktop dashboard
• Described by AWS Distinguished Developer Advocate Darko Mesaros as: “a persistent, open-source development workspace for work that is bigger than a single task in a single session”

Origins and deployment

• Originally internal Amazon project MeshClaw; adopted by more than 39,000 Amazon builders in under six months
• Can deploy entirely inside customer environments (laptops, containers, VMs) without requiring an AWS account or AWS-managed control plane
• Reference apps: DevFleets (worktree management), Issue Radar (issue/PR triage), Task Runner (long-running tasks)

Open source caveats

• Built around open standards (ACP, MCP)
• At launch runs on the proprietary Kiro CLI (metered by credits) — orchestration layer is open; day-one harness wiring is Kiro-specific
• Existing .kiro configs (steering, skills, custom agents) carry over for Kiro users
• Crew itself free; customers pay for connected agents/tools
• Public steering committee / open governance model planned; Kiro and AWS engineers initial maintainers

GitHub: kirodotdev/KiroCrew (Apache 2.0); docs at kiro.dev/docs/crew/