architecture-expert

Formalize architecture constraints and trade-offs into ADR-lite decision records.

9|Updated Aug 20, 2025
One-click install
npx skills add https://github.com/mae616/ai-template --skill architecture-expert
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: architecture-expert
Source: https://github.com/mae616/ai-template/tree/main/.claude/skills/architecture-expert
Command: npx skills add https://github.com/mae616/ai-template --skill architecture-expert

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill translates high-level architectural intent into concrete constraints, boundaries, and trade-offs, enabling ADR-lite style decisions to be captured and reused.

Core Features & Use Cases

  • Constraint-driven design: Define component boundaries, ownership, interfaces, data flows, and non-functional requirements (scalability, reliability, observability) to guide implementation.
  • ADR-lite decision records: Propose options, compare trade-offs, and document rationale to support future changes.
  • Guided collaboration: Facilitate human-AI collaboration for architecture exploration, trade-off analysis, and progressive refinement.

Quick Start

Provide the system context (purpose, constraints, and non-functional requirements). The AI will outline component boundaries, data ownership, interfaces, and ADR-lite decisions, and suggest the next actions.

Frequently Asked Questions about architecture-expert

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

FAQPage Schema
How do I document software architecture trade-offs and constraints for my system design?

You document software architecture trade-offs by capturing component boundaries, data ownership, interfaces, and non-functional requirements into ADR-lite records. This translates high-level architectural intent into concrete constraints with traceable rationale and validation steps.

What is the best way to define component boundaries and data ownership in a distributed system?

The best way to define component boundaries and data ownership is through constraint-driven design. By applying non-functional requirements like scalability and reliability, you can outline clear interfaces, dependencies, and data flows to guide implementation and capture ADR-lite decisions.

How does ADR-lite documentation work for guiding architecture decisions?

ADR-lite documentation works by proposing options, comparing trade-offs, and documenting rationale for architecture decisions. It formalizes non-functional requirements and system boundaries to ensure decisions are traceable and reusable for future system modifications.

Do I need to provide non-functional requirements to generate system architecture boundaries?

Yes, you need to provide non-functional requirements like scalability, reliability, and observability as system context. Providing these constraints allows the architecture design process to outline accurate component boundaries, data ownership, interfaces, and ADR-lite decisions.

Can I use this approach to analyze architecture trade-offs for progressive refinement?

Yes, you can use this approach to facilitate human-AI collaboration for architecture exploration and trade-off analysis. It supports progressive refinement by presenting options, comparing trade-offs, and documenting rationale to guide iterative system design improvements.