subagent-driven-development

Read the plan once, extract all tasks with full text into a TodoWrite list, then dispatch a fresh implementer subagent per task with the complete task text and context. After each task, run a spec compliance review followed by a code quality review before marking it complete.

Subagent-driven development runs in the same session with fresh subagents per task and no human-in-the-loop between tasks. Executing plans runs in a parallel session with a handoff, which suits cases where you want to leave the current session.

Avoid it when tasks are tightly coupled and cannot be implemented independently, when there is no written implementation plan, or when you prefer a parallel session workflow. Also never dispatch multiple implementation subagents in parallel because they conflict.

Assess the blocker: provide more context and re-dispatch if it is a context problem, use a more capable model if it needs more reasoning, break the task into smaller pieces if too large, or escalate to the human if the plan itself is wrong. Never force the same model to retry unchanged.

Spec compliance verifies the implementer built exactly what was requested, nothing more or less, by reading the actual code rather than trusting the report. Reviewing code quality before correctness wastes effort on code that may need to change to meet the spec.