subagent-driven-development

Extract all tasks from the plan with full text, create a todo list, then dispatch one 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 stays in the same session with fresh subagents per task and no human-in-loop between tasks, enabling faster iteration. Executing plans uses a parallel session with a context switch and handoff between sessions.

Avoid it when tasks are tightly coupled and cannot run independently, when no implementation plan exists yet, or when you intend to work in a separate parallel session. In those cases brainstorm first or use the executing-plans workflow instead.

Assess the blocker: provide more context and re-dispatch for context problems, switch to a more capable model for reasoning-heavy tasks, split oversized tasks into smaller pieces, or escalate to the human if the plan itself is wrong. Never force the same model to retry unchanged.

Use a fast cheap model for mechanical tasks touching one or two files with clear specs, a standard model for multi-file integration work, and the most capable model for architecture, design, and review tasks.

Spec compliance confirms the implementation matches requirements with nothing missing or extra, which must hold before judging code quality. Reviewing quality first wastes effort on code that may need rework to meet the spec.