design-advisor

Coordinate multi-agent critiques, plans, or reviews of software architecture.

4|Updated Jan 22, 2026
One-click install
npx skills add https://github.com/rodekruis/qualitative-feedback-analysis --skill design-advisor-rodekruis
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: design-advisor
Source: https://github.com/rodekruis/qualitative-feedback-analysis/tree/main/.claude/skills/design-advisor
Command: npx skills add https://github.com/rodekruis/qualitative-feedback-analysis --skill design-advisor-rodekruis

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps prevent costly mistakes by providing expert architectural advice before any code is written, ensuring designs are robust, user-friendly, and free from over-engineering.

Core Features & Use Cases

  • Architecture Critique: Identifies potential issues in code structure, deployment, API design, domain correctness, and complexity.
  • Solution Planning: Designs new features or systems based on defined requirements, considering multiple expert perspectives.
  • Change Review: Evaluates proposed modifications to existing architecture.
  • Use Case: Before starting a new microservice, use plan: to get a consensus design from an architect, UX advocate, domain expert, and devil's advocate.

Quick Start

Use the design-advisor skill to critique the current architecture for the feedback analysis backend.

Frequently Asked Questions about design-advisor

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

FAQPage Schema
How do I critique software architecture before starting implementation?

To critique software architecture, you can use an agent team to evaluate your design from four expert perspectives: clean hexagonal architecture, API ergonomics, domain expertise, and overengineering prevention. This provides consensus recommendations before implementation.

What is the best way to plan a new microservice architecture?

The best way to plan a new microservice architecture is to define your requirements and run a planning mode analysis that designs the system from multiple expert perspectives. This yields a robust, user-friendly consensus design free from overengineering before code is written.

How does a multi-agent review system work for system design?

A multi-agent review system for system design works by deploying specialized agents to independently evaluate architecture from distinct angles, such as API ergonomics and domain expertise. They then combine their findings into a unified consensus recommendation.

Can I use architectural critique to prevent overengineering in my codebase?

Yes, you can use architectural critique to prevent overengineering by including a dedicated devil's advocate perspective in your review process. This specifically targets unnecessary complexity in code structure, deployment, and API design.

When do I need to review proposed changes to an existing architecture?

You need to review proposed changes to an existing architecture when you are evaluating modifications to ensure they do not break domain correctness or API ergonomics. A structured change review evaluates these shifts from multiple expert perspectives.