Architecture Decision Framework

What your AI can do with it

• Asks discovery questions to classify the project before designing
• Evaluates trade-offs between architecture options
• Records decisions as ADRs using provided templates
• Selects patterns using decision trees and anti-pattern guidance
• References MVP, SaaS, and Enterprise example architectures
• Runs a validation checklist before finalizing architecture

Getting started

1. Have an agent environment that supports loading skills.
2. Add the architecture skill to your available skills.
3. Ask the agent to make an architecture decision or analyze a system design; it will ask discovery questions, weigh trade-offs, and write ADRs.
4. Optionally pair it with related skills such as database-design, api-patterns, or deployment-procedures.

What this skill tells your AI

The instructions your AI receives, as published by davila7/claude-code-templates in cli-tool/components/skills/development/architecture/SKILL.md and read by ahel’s review.

Signals

GitHub stars

32k

Forks

4k

Last commit

Oct 2026

Others that do the same job

Questions

What does the skill read?

It only reads project files and its own reference documents, covering context discovery, trade-off analysis, pattern selection, examples, and a patterns reference. It reads only files relevant to the request.

When should I use it?

Use it when making architecture decisions or analyzing system design, such as starting an architecture design or documenting decisions.

Does it modify my project?

Its allowed tools are Read, Glob, and Grep, so it reads files rather than modifying them; it records choices as ADRs using templates.

What principle does it follow?

Start simple and add complexity only when proven necessary, since removing complexity is much harder than adding it.