Amazon has introduced Kiro Crew, an open source framework designed to manage multiple Kiro coding agents across different sessions, tools, and tasks. The system was announced as part of the company’s ongoing push to expand asynchronous automation in software development.
The new workspace enables developers to assign coding tasks to AI agents that operate without continuous human supervision. These tasks include incident investigation, ticket triage, code migrations, and pull request monitoring.
Purpose and Design
Kiro Crew is built to coordinate several agents that work independently on assigned tasks, potentially over extended periods. The framework is designed to handle workloads that do not require immediate human intervention, allowing teams to focus on higher priority activities while agents manage routine or background processes.
According to the announcement, Kiro Crew supports parallel execution of tasks, with agents capable of interacting with multiple tools and repositories. The open source nature of the project means developers can inspect, modify, and deploy the system within their own environments.
Technical Foundations
Kiro Crew is built primarily in Python and relies on the Kiro agent architecture, which is Amazon’s internal coding assistant for developer workflows. The framework uses a task definition schema that specifies inputs, outputs, and constraints, enabling reproducible and observable agent executions.
The system includes a command line interface for creating, listing, and inspecting tasks. It also provides hooks for integrating with popular version control platforms, issue trackers, and communication tools. Amazon has published documentation and example configurations to help teams adopt the framework.
Context and Significance
The introduction of Kiro Crew aligns with broader industry trends toward multi agent and asynchronous AI systems. Many organizations are testing ways to delegate repetitive or well defined coding work to automated agents while maintaining oversight and quality control.
Amazon previously released Kiro as an experimental open source project in late 2025. Kiro Crew expands that vision by addressing coordination, scheduling, and fault tolerance for multiple concurrent agents.
Early users and developers have noted that asynchronous agent systems could reduce cycle times for maintenance tasks and free up engineering capacity. However, adoption remains early, and questions about reliability, security, and governance still need to be addressed in production environments.
Next Steps
Amazon plans to continue developing Kiro Crew with contributions from the open source community. The project is currently available on GitHub under a permissive license, and the company encourages feedback and experimentation.
Developers interested in using Kiro Crew can access the repository and explore configuration options for their own workflows. Future updates are expected to add more integrations, improved scheduling algorithms, and enhanced observability features.
