What problem does it solve?
This Skill helps you execute an implementation plan reliably by dispatching fresh subagents per task and enforcing a two-stage review (spec compliance first, then code quality) to prevent errors and scope drift.
Core Features & Use Cases
- Per-task fresh implementation: Runs an implementer subagent for each task with the full task context so work stays focused and avoids accumulated confusion.
- Two-stage automated review: Performs a spec compliance review followed by a code quality review, looping back to fix issues until approved.
- TDD-first execution: Ensures implementer subagents follow a test-driven development flow (write failing tests, implement minimally, re-run tests, verify no regressions).
- Final integration review: After all tasks are completed, dispatches a reviewer subagent to check consistency and integration across the full change set.
Use case example: You have a multi-step implementation plan for a feature (models, endpoints, utilities, tests) and want each task to be implemented quickly while still guaranteeing it matches the plan and remains high-quality after review.
Quick Start
Use subagent-driven-development to execute every task in your implementation plan with an implementer subagent, then run spec compliance and code quality reviews until each task is approved.
