creating-adrs

Generate MADR-format Architecture Decision Records through a six-phase workflow.

11|1|Updated Aug 6, 2025
One-click install
npx skills add https://github.com/thkt/claude-config --skill creating-adrs
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: creating-adrs
Source: https://github.com/thkt/claude-config/tree/main/ja/skills/creating-adrs
Command: npx skills add https://github.com/thkt/claude-config --skill creating-adrs

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

MADR形式のArchitecture Decision Recordsを、検証済みの6段階プロセスで自動生成します。重複ADRの検出、テンプレート選択、参照収集、校正、インデックス更新、復旧策を統合的にサポートします。

Core Features & Use Cases

  • テンプレート自動選択: technology-selection、architecture-pattern、process-change、deprecation の4テンプレートから適切な ADR テンプレートを自動選択します。
  • 重複 ADR の検知: 類似 ADR の検出と提案で、組織内の決定の重複を避けます。
  • インデックス更新: docs/adr/README.md の ADR 採番・状態反映を自動で更新します。
  • 多言語対応ガイダンス: 日本語/英語の併記サポートで、グローバルチームの協働を促進します。

Quick Start

/adr "Adopt Zustand for State Management" を実行すると、ADR生成の6段階ワークフローが開始されます。テンプレートの選択、参照の収集、検証、インデックス更新まで自動で進行します。

Frequently Asked Questions about creating-adrs

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

FAQPage Schema
How do I automate Architecture Decision Record creation in MADR format?

MADR-format ADR automation uses a six-phase workflow to generate structured decision records with template selection, duplicate detection, reference collection, validation, index updates, and error recovery, eliminating manual drafting errors across software projects.

Can ADR templates be automatically selected based on decision type?

Yes, ADR template auto-selection chooses from four predefined templates—technology-selection, architecture-pattern, process-change, and deprecation—matching the decision category to enforce consistent documentation structure.

How does duplicate ADR detection prevent redundant architectural decisions?

Duplicate ADR detection identifies and flags similar existing records, suggesting consolidated decisions to avoid organizational redundancy and maintain a single source of truth for architectural choices.

Does this support multilingual ADR documentation for global teams?

Yes, multilingual guidance supports parallel Japanese and English documentation, enabling global teams to collaborate while maintaining architectural decision records in their preferred languages.

What happens to the ADR index when new records are created?

Index auto-maintenance automatically updates docs/adr/README.md with ADR numbering and status reflections, keeping the central registry synchronized with newly created or modified decision records.

Can ADR generation recover from errors during the workflow?

Yes, integrated error recovery mechanisms safeguard the six-phase ADR generation process, allowing the workflow to handle failures and maintain data integrity throughout document creation and indexing.