What problem does it solve? Executing a multi-task implementation plan in a single AI session often leads to context pollution, skipped reviews, and inconsistent quality. This Skill orchestrates plan execution by dispatching a fresh subagent per task, keeping the controller's context clean and enforcing review gates after every task. ## Core Features & Use Cases - Fresh Subagent Per Task: Each task runs in an isolated subagent with curated context, preventing cross-task confusion and preserving the controller's coordination capacity. - Two-Stage Review Gates: Every task passes a spec compliance review first, then a code quality review, with fix-and-re-review loops until approved. - Status-Based Escalation Handling: Implementer subagents report DONE, DONE_WITH_CONCERNS, NEEDS_CONTEXT, or BLOCKED, and the controller responds with targeted remediation such as more context, a more capable model, or task decomposition. - Use Case: Given a written implementation plan with five independent tasks, execute all of them in the current session with automatic spec and quality reviews after each, then run a final whole-implementation review before finishing the branch. ## Quick Start Use subagent-driven development to execute the implementation plan in docs/plans/feature-plan.md task by task with reviews after each task.
