subagent-driven-development

Autonomous plan execution is handled by dispatching independent subagents for each task, followed by a mandatory two-stage review process for spec compliance and code quality. It manages task delegation and iterative fixing within a single session.

Subagent-driven development is a technique where independent subagents implement discrete tasks from a plan. It handles code review through a mandatory two-stage process: first checking spec compliance, then evaluating code quality, with iterative feedback loops for fixes.

You delegate development tasks by providing a detailed implementation plan. The Skill breaks down the plan into discrete tasks, assigning each to a fresh subagent, and facilitates quick feedback loops where implementer subagents fix issues identified by reviewers.

Yes, this task delegation approach suits complex development workflows by managing task delegation, question handling, iterative fixing, and final code review within a single session, ensuring rigorous review and high-quality iterative development.

Automated plan execution requires a detailed implementation plan file, such as a markdown document located at a specified path. The Skill uses this plan to break down work into discrete tasks assigned to independent subagents.