What problem does it solve? Turning a feature spec or rough idea into an executable implementation plan is error-prone: tasks get mis-sized, dependencies get missed, and requirements silently drop out. This Skill produces a rigorous, task-by-task TDD plan where every task has exact verification commands, acceptance proofs, and explicit dependency ordering. ## Core Features & Use Cases - Multi-agent architecture analysis: Dispatches Architect, Tech Lead, and QA Engineer sub-agents sequentially to map component boundaries, select patterns and libraries, and assess testability before any task is written. - TDD task decomposition: Synthesizes agent reports into ordered tasks following RED-GREEN-Verify-Commit, with complexity ratings, failure strategies, and file-limit enforcement. - Independent plan review: A plan-reviewer agent validates spec coverage, dependency correctness, acceptance-proof executability, and fidelity to user-provided design artifacts. - Use Case: After approving a brainstorm spec for a new billing feature, run the plan skill to get a dated plan document in docs/specs/ with 5-10 milestone tasks, each carrying exact test commands and per-AC proof blocks ready for the execute phase. ## Quick Start Ask the AI to run the plan skill on your approved spec at docs/specs/my-feature-spec.md to generate a TDD implementation plan.