subagent-driven-development

Subagent-driven development improves iterative development by dispatching independent subagents for each task and enforcing a mandatory two-stage review process. This ensures both spec compliance and code quality before moving forward.

You can execute a complex implementation plan by pointing the Skill to your plan file. It dispatches a fresh subagent for each task, manages iterative fixes, and answers questions within a single session to ensure fast, high-quality code iteration.

The two-stage review process first checks for spec compliance to verify the task meets all requirements, then evaluates code quality. This mandatory sequence ensures high-quality, reliable iteration by catching issues early through automated reviews and clear feedback loops.

Yes, you can manage task execution and iterative fixes within a single session. The Skill handles the entire implementation by dispatching independent subagents, managing task completion, and facilitating rapid iteration without requiring manual session restarts.

No, you do not need to manually break down implementation plans. The Skill streamlines the development process by automatically decomposing the implementation plan into independent tasks, each handled by a fresh subagent with rigorous two-stage reviews.