subagent-driven-development

Execute a multi-task implementation plan by dispatching one fresh subagent per task with isolated context, then applying a two-stage review (spec compliance and code quality) before advancing. Use when you have a written plan with 3+ tasks, when tasks would otherwise cause context bleed, when you need reproducible per-task output, or when review gates are non-negotiable. Covers per-task briefing templates, the two-stage review rubric, structured hand-off between dependent tasks, and anti-patterns like shared long-context drift and skipped reviews. Complements dispatching-parallel-agents (for fan-out work) and executing-plans. Triggers: 'execute this plan', 'work through these tasks', 'dispatch per task', 'one-task-at-a-time', 'review-gated implementation'.