What problem does it solve? Plans written for AI-driven development often contain gaps: missing requirements, broken task dependencies, dangling task IDs, or vague acceptance criteria. This Skill inspects and refines a task plan before any code is written, catching structural and coverage problems early so execution phases run without rework. ## Core Features & Use Cases - Gap Analysis: Compares PLAN.md against SPEC.md and codebase patterns, classifying findings as Must fix, Should fix, or Observations. - Dependency Validation: Detects topological violations, dependency cycles, dangling T-NNN references, and mermaid graph mismatches, then renumbers tasks atomically. - State Management: Advances the SPAE workstream state from inspect to build, rebuilding the task registry in STATE.json with verified write-then-read persistence. - Use Case: After generating a plan from a specification, run this Skill to verify every requirement maps to a task, every task has testable acceptance criteria and an actionable intent, and the task graph is topologically ordered before build begins. ## Quick Start Run the inspect phase on the current SPAE workstream to validate and refine PLAN.md against SPEC.md before starting the build phase.
