subagent-driven-development

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

Subagent-driven development stays in the same session with no context switch, uses a fresh subagent per task, and iterates faster without human-in-the-loop between tasks. Executing plans in a parallel session involves a handoff to a separate session.

Use the least powerful model that can handle each role. Mechanical tasks touching one or two files with a clear spec fit cheap models, multi-file integration tasks fit standard models, and architecture or review tasks need the most capable model.

Assess the blocker: provide more context and re-dispatch if it is a context problem, re-dispatch with 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.

Avoid it when there is no implementation plan yet, when tasks are tightly coupled and cannot run independently, or when you need a parallel session instead of same-session execution. Never dispatch multiple implementation subagents in parallel since they conflict.