subagent-driven-development

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

Stage one is spec compliance review, verifying the implementation matches the original task spec with no scope creep. Stage two is code quality review, checking conventions, error handling, test coverage, and security. Spec review must pass before quality review begins.

Fresh subagents prevent context pollution from accumulated state, giving each task clean focused context without confusion from prior tasks' code or reasoning. The trade-off is more subagent invocations, but issues are caught early rather than compounding.

Each task should represent 2-5 minutes of focused work, such as creating a single model or endpoint. Broad tasks like implementing an entire authentication system should be split into discrete units like password hashing, token generation, and registration endpoints.

The implementer subagent or a new fix subagent addresses the specific issues, then the reviewer re-reviews. This loop repeats until approved, and you never proceed to the next task while either review has open critical or important issues.

Avoid it when tasks touch the same files and cannot run independently, when no implementation plan exists, or when you are tempted to skip reviews. Starting without a plan or accepting close-enough spec compliance are explicit anti-patterns.