subagent-driven-development

Automate code implementation and reviews by dispatching independent subagents to execute tasks, followed by a two-stage review for spec compliance and code quality. This streamlines development by orchestrating specialized agents for implementation and automated quality assurance checks.

Subagent-driven development is a workflow that executes implementation plans by dispatching independent subagents for each task. It ensures code quality through a two-stage review process, applying specialized agents for spec compliance and code quality assessment to enforce best practices.

Execute an implementation plan step-by-step by dispatching specialized subagents for each decomposed task. The process includes automated implementation followed by iterative two-stage reviews for spec compliance and code quality, handling fixes within the same session until the feature is complete.

Subagent orchestration supports TDD and automated code review by dispatching specialized subagents for implementation, spec review, and code quality assessment. The two-stage review process enforces spec compliance and best practices, ensuring adherence to requirements.

Break down complex feature plans into manageable, independent tasks using task decomposition. The workflow then dispatches specialized subagents for each task, autonomously implementing each part and iterating on fixes through multi-stage reviews until ready for merging.

AI subagents for software development require a detailed feature plan to function effectively. The workflow is limited to executing provided plans and performing two-stage reviews for spec compliance and code quality, rather than autonomously designing software architecture or generating the initial implementation plan.