subagent-driven-development

Executing a complex implementation plan autonomously involves breaking it into independent tasks and dispatching a dedicated subagent for each task. The subagent-driven-development process ensures strict adherence to predefined specifications within a single agent session.

Subagent dispatch for iterative development assigns each decomposed task to a fresh, independent subagent for focused implementation. This is followed by a two-stage review process for spec compliance and code quality, ensuring rapid iteration without manual oversight.

Automating code review during agent execution is best handled through a two-stage review process using dedicated reviewer subagents. This validates both spec compliance and code quality, applying automated testing and multi-stage validation loops to ensure high-quality output.

Yes, autonomous task execution requires a detailed feature implementation plan as input. The skill automatically breaks down this plan into manageable, independent tasks, assigning each to a fresh subagent for focused implementation and rigorous review.

You should not use subagents for development when your scenario lacks predefined specifications or requires manual oversight between tasks. This approach is designed for autonomous task execution, iterative development, and strict adherence to predefined specifications within a single agent session.