What problem does it solve? Running a full review gauntlet on a plan normally means answering 15-30 intermediate questions across CEO, design, and engineering reviews. This Skill runs all three review phases automatically, deciding routine questions with 6 decision principles and only surfacing genuine taste decisions at one final approval gate. ## Core Features & Use Cases - Sequential Review Pipeline: Executes CEO, design, and eng review phases in strict order, each at full depth, reading the actual review skill files from disk. - Auto-Decision Engine: Applies 6 principles (completeness, boil lakes, pragmatic, DRY, explicit over clever, bias toward action) to answer intermediate questions, with a full audit trail of every decision. - Final Approval Gate: Surfaces taste decisions (close approaches, borderline scope, codex disagreements) and User Challenges where both models disagree with your stated direction. - Restore Point: Captures the plan file's original state before any changes so you can roll back and re-run. - Use Case: You have a rough plan file for a new feature. Run /autoplan and get back a fully reviewed plan with scope, architecture, and UX decisions made, plus a short list of judgment calls to approve. ## Quick Start Ask the AI to run autoplan on your current plan file to execute the full CEO, design, and engineering review pipeline automatically.
