# Architecture

> Judges the high-level shape of the solution — the architectural pattern, structure, placement, and data model.

- Skill: `elliottlawson/architecture` (Agent Skill)
- Install (CLI): `npx skillmds@latest add elliottlawson/architecture`
- Raw SKILL.md: https://api.skillmd.com/api/skills/elliottlawson/architecture/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- License: MIT
- Author: elliottlawson (https://skillmd.com/u/elliottlawson)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/elliottlawson/architecture

---


# Architecture

Judge the solution at the system level, not the code level.

- **Architectural pattern.** What pattern is the work built on — and is it the right one for this problem?
- **Decomposition.** How is the work split into pieces? Are the pieces in the right modules, layers, or services?
- **Dependency direction.** Do the pieces depend on each other the right way, or is something coupled that should be separate?
- **Data model.** Is the schema and migration shape right? How does state flow through the change?
- **Right-sized.** Over-built for what it needs, or too thin to hold up?

A decision that would ripple across the codebase is architecture. Localized details — how the framework is used, conventions, tests, naming — are judged by the implementation and craft passes, not here.

If the shape is wrong, say so first. Nothing else matters until the shape is right.