skills.rest
Marunion
MarunionCOMMUNITY
Explore all skills by this author →
GitHub

arch

Facilitate multi-agent debates to derive software architecture designs.

Updated Mar 7, 2026
Search Tags:#design#architecture#documentation#frontend#software development#backend#multi-agent
One-click install
npx skills add https://github.com/Marunion/ai-factbomb --skill arch-marunion
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: arch
Source: https://github.com/Marunion/ai-factbomb/tree/main/archflow-main/skills/arch
Command: npx skills add https://github.com/Marunion/ai-factbomb --skill arch-marunion

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill automates the complex process of software architecture design by facilitating a structured debate between specialized AI agents, ensuring optimal solutions are derived from both project-specific context and industry best practices.

Core Features & Use Cases

  • Multi-Agent Debate: Domain Architect and Best Practice Advisor agents collaborate in rounds to refine design proposals.
  • Contextual Design: Considers existing project structure, technical constraints, and feature requirements.
  • Best Practice Integration: Incorporates industry standards for scalability, maintainability, and robustness.
  • Use Case: When designing a new backend service, this Skill can generate a comprehensive architecture document, including tech stack, database schema, API specifications, and implementation plans, by analyzing requirements and debating design choices.

Quick Start

Use the arch skill to design the backend architecture for the 'user-auth' service based on its spec.md file.

Frequently Asked Questions about arch

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

FAQPage Schema
How does multi-agent debate improve software architecture design?▼

Multi-agent debate improves software architecture design by structuring rounds of collaboration between a Domain Architect and a Best Practice Advisor. This refines design proposals by synthesizing project-specific context with industry standards for scalability and maintainability.

How do I generate backend architecture documentation from a spec file?▼

You generate backend architecture documentation by providing a spec file to the Skill. It analyzes requirements, explores project context, and synthesizes a comprehensive design document including tech stack, database schema, and API specifications.

Can I use this to design frontend architecture for an existing project?▼

Yes, you can design frontend architecture for existing projects. The Skill evaluates current project structures and technical constraints to derive optimal design proposals, ensuring new frontend features integrate with established codebases.

Do I need a specific model to generate comprehensive design proposals?▼

Yes, you need the Opus model to generate comprehensive design proposals. The Skill requires this advanced model to facilitate the multi-agent debate effectively and ensure high-quality output for complex software development scenarios.

What is the best way to automate API specifications and database schema creation?▼

The best way to automate API specifications and database schema creation is using a structured multi-agent debate. It evaluates feature requirements against best practices, outputting complete implementation plans and architecture documents.

Related Skills

Explore More Software Engineering Skills→
fmbetancourtfmbetancourt

speckit-specify

Generate structured feature specifications from natural language inputs.

Community
Advanced
AdawoduAdawodu

engineering-backend-architect

Design backend system architecture specifications for microservices, APIs, and database schemas.

Community
Advanced
markrammmarkramm
3

software-kb

Coordinate software knowledge-base item flow through Pyrite kanban cycles.

Community
Intermediate
tmalcolm-0607tmalcolm-0607

debate

Run structured multi-agent adversarial debates with configurable formats and verdict generation.

Community
Advanced