multi-agent-review-plan

Coordinate parallel Codex and Antigravity reviews of high-level technical plans.

Updated Jul 8, 2026

Search Tags:#planning #architecture #quality assurance #code review #software design #multi-agent

Ready to run

npx skills add https://github.com/missdeer/skills --skill multi-agent-review-plan

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: multi-agent-review-plan
Source: https://github.com/missdeer/skills/tree/main/multi-agent-review-plan
Command: npx skills add https://github.com/missdeer/skills --skill multi-agent-review-plan

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill prevents implementation errors and architectural misalignment by subjecting high-level project plans to a rigorous, dual-agent review process before any code is written.

Core Features & Use Cases

Dual-Agent Review: Orchestrates parallel reviews by Codex and Antigravity to ensure architectural and business logic soundness.
Hard Scope Enforcement: Strictly filters out implementation details, focusing only on technology selection, architecture, business flow, and logic.
Closed-Loop Iteration: Aggregates feedback, manages dismissed items, and iterates until the plan is robust and aligned with project goals.

Quick Start

Invoke the multi-agent-review-plan skill to perform a dual-reviewer audit on the current high-level plan before starting implementation.

Frequently Asked Questions about multi-agent-review-plan

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

FAQPage Schema

How do I review a high-level technical plan before coding begins?▼

High-level technical plan review is facilitated by a dual-agent process that coordinates parallel feedback on architecture and business logic to ensure alignment before implementation. This structured approach aggregates iterative feedback to minimize technical debt.

What is multi-agent review for software design and architecture?▼

Multi-agent review for software design is a structured process where specialized agents provide parallel feedback on technology selection, architecture, and business flow. It enforces strict scope boundaries to exclude implementation-level details and refines project direction through closed-loop iteration.

How do I prevent architectural misalignment and technical debt in project planning?▼

Preventing architectural misalignment requires subjecting high-level project plans to a rigorous dual-agent review process before any code is written. By auditing technology selection and business logic soundness, the plan is iteratively refined to minimize technical debt.

Does the plan review process exclude implementation details?▼

The plan review process strictly filters out implementation details through hard scope enforcement, focusing only on technology selection, architecture, business flow, and logic. This ensures the review remains high-level and does not dive into coding specifics.

Can I use automated agents to ensure business logic soundness in software design?▼

Automated agents can ensure business logic soundness by orchestrating parallel reviews focused on architectural and business alignment. The process aggregates feedback, manages dismissed items, and iterates until the high-level plan robustly aligns with project goals.

When do I need a dual-agent audit for my software architecture plan?▼

A dual-agent audit for software architecture is needed when you have a defined high-level plan and want to prevent implementation errors before coding begins. It is essential for refining project direction and ensuring strict architectural and business alignment.