subagent-driven-development

To manage plan execution, this skill dispatches a fresh subagent for each task to keep context isolated and reduce cross-task interference. It coordinates independent task execution within a single session while maintaining transparent progress tracking.

A two-stage review workflow verifies task progress through spec compliance first, then code quality. This automated review mechanism ensures the output meets functional specifications before evaluating the structural integrity of the code itself.

You automate task dispatch and iterative refinement by orchestrating multi-task plans with automatic progress tracking and rework loops. The system dispatches a fresh subagent per task and enforces reviews after each completion.

Subagents keep context clean by isolating each task within a fresh execution environment. This per-task isolation prevents cross-task interference and maintains transparent progress throughout the end-to-end plan execution.

Yes, automated spec compliance review works with TDD planning by verifying that task outputs meet formal specifications before running code quality checks. This iterative refinement supports complex, parallelizable workflows.

You should avoid using autonomous subagents for task orchestration when tasks have heavy interdependencies that prevent isolated execution, as the system relies on per-task isolation and independent execution within a single session.