architect-init

Detect architectural patterns and generate ADR drafts from brownfield codebases.

Updated Apr 21, 2026
One-click install
npx skills add https://github.com/chaimt/GarminShabbatMode --skill architect-init
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: architect-init
Source: https://github.com/chaimt/GarminShabbatMode/tree/main/.cursor/skills/architect-init
Command: npx skills add https://github.com/chaimt/GarminShabbatMode --skill architect-init

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Brownfield codebases often lack explicit architectural decisions, leaving teams to manually infer structure and relationships. This skill automates discovery and documents inferred ADRs to provide a clear architectural record.

Core Features & Use Cases

  • Detect architectural patterns (monolith, microservices, modular monolith) from an existing codebase.
  • Generate ADR drafts organized by sub-systems and prepare for clarification.
  • Identify information gaps and pose clarifying questions to stakeholders.

Quick Start

Run architect-init in your brownfield repository to generate initial ADR drafts and a first-pass architectural map.

Frequently Asked Questions about architect-init

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

FAQPage Schema
How do I document inferred architecture from a brownfield codebase?

To document inferred architecture from a brownfield codebase, run the analysis to generate ADR drafts organized by sub-systems. The process identifies architectural patterns and cites concrete code evidence to assign a confidence level for each inference.

What is the best way to generate ADRs for an existing codebase?

Generating ADRs for an existing codebase involves detecting architectural patterns like monolith or microservices and drafting records per sub-system. It links to clarifications for validation and cites concrete files, directories, and patterns as evidence.

How do I identify architectural patterns in a legacy codebase?

Identifying architectural patterns in a legacy codebase requires analyzing the directory structure and code relationships. The skill detects patterns such as monolith, microservices, or modular monolith and documents them as inferred architectural decisions.

Can I use code analysis to prepare architectural clarification questions for stakeholders?

Yes, you can use code analysis to prepare architectural clarification questions. The process identifies information gaps in the inferred architecture and generates targeted questions for stakeholders to validate the assigned confidence levels of each decision.

Does this approach work for both monolith and microservices architecture documentation?

Yes, this approach works for both monolith and microservices architecture documentation. It detects various architectural patterns from the existing codebase and generates ADR drafts organized by sub-systems regardless of the initial structure.

What are the limitations of inferring architectural decisions from code without explicit documentation?

Inferring architectural decisions from undocumented code relies on analyzing structural patterns, which means inferences may have information gaps. It assigns confidence levels to each inference and generates clarifying questions to resolve ambiguity with stakeholders.