plan-review

Coordinate five expert agents to review PLAN phase design documents.

8|2|Updated Oct 28, 2025
One-click install
npx skills add https://github.com/morodomi/tdd-skills --skill plan-review-morodomi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: plan-review
Source: https://github.com/morodomi/tdd-skills/tree/main/plugins/tdd-core/skills/plan-review
Command: npx skills add https://github.com/morodomi/tdd-skills --skill plan-review-morodomi

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Plan Review enables parallel, multi-perspective evaluation of PLAN phase design by coordinating five expert agents to speed up feedback and improve quality.

Core Features & Use Cases

  • Parallel expert reviews: five agents assess design policies, cycle docs, test lists, and planned file changes concurrently.
  • Automated result synthesis: scores, PASS/WARN/BLOCK decisions, and consolidated recommendations.
  • Use Case: A product team uses PLAN review to verify cycle docs and changes before RED/GREEN steps, reducing drift and rework.

Quick Start

To run a PLAN review, load the cycle doc, start the five reviewers, and collect the consolidated score before proceeding to RED.

Frequently Asked Questions about plan-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run a parallel design review for cycle documents?

Parallel plan reviews use five expert agents to evaluate design policies, cycle docs, test lists, and planned file changes concurrently, producing a 0-100 confidence score and a PASS, WARN, or BLOCK decision.

How does automated result synthesis work for multi-agent plan reviews?

Automated synthesis consolidates outputs from five concurrent expert agents into a 0-100 confidence score and a final PASS, WARN, or BLOCK decision to reduce manual review effort.

Can I use parallel expert reviews to check planned file changes before coding?

Yes, parallel expert reviews target design reviews of cycle documents and planned file changes before RED or GREEN steps, reducing drift and rework.

What is a 0-100 confidence score in a plan review?

A 0-100 confidence score is the quantitative metric produced by synthesizing parallel expert agent outputs, determining the final PASS, WARN, or BLOCK decision for the plan.

Do I need reference materials to start a multi-agent plan review?

You need to load cycle documents and reference materials to start the five reviewers, which then output a consolidated score before you proceed to the next phase.

When should I use an automated plan review over manual design checks?

Use automated plan review when you need to speed up multi-perspective evaluation of cycle documents before implementation, replacing slow manual checks with concurrent expert synthesis.