By kicchann
Git-native要件管理ツールReqordの開発支援プラグイン。要件から仕様、実装、検証までのトレーサビリティを維持しながら、要件作成・設計・検証・フィードバック運用を支援する。
Reqordの要件・仕様に基づく設計エージェント。 design.mdとProjectContextの設計原則に従った設計を行う。
Reqordの要件・仕様を踏まえたコードベース調査エージェント。 design.mdのコンポーネント構成と実コードの照合、実装状況の分析を行う。
Reqordの要件・仕様に基づくコードレビューエージェント。 Requirementのsuccess criteriaに対する実装の網羅性とdesign.mdとの一致をチェックする。
spec/req/issueの包括的コンテキストを一括表示する。実装着手前の状況把握に使用。
Reqordのデータモデル(Requirement, Specification, ProjectContext, Feedback)とCLI操作ルールの共通知識ベース。
reqordデータ(Requirement, Specification, ProjectContext)の編集・改善を統合的に行う。reqのSMART品質改善、specのdesign.md技術設計書の生成・更新、contextの充実化をカバーする。
フィードバック運用(作成・同期・分類・リンク・クローズ・フラグ解消)。GitHub Issueの作成または既存フィードバックの取り込みワークフローを実行する。
reqordデータ(Requirement, Specification)の新規作成。対話的にデータを作成し、初期品質を確保する。
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Structured context for the AI era, version-controlled alongside your code.
Reqord stores requirements as structured data (YAML + Markdown) inside your Git repository, giving every requirement a clear lifecycle from draft to implementation. No SaaS, no backend -- just git clone and your entire requirements history is there, version-controlled alongside your code, ready for humans and AI tools alike.
Software teams lose track of requirements. The consequences compound over time:
Reqord solves these by making requirements structured, versioned, and traceable -- stored right where your code lives.
Reqord enforces a 3-layer traceability model that connects intent to implementation:
Requirement (What) --> Specification (How) --> GitHub Issue (Tasks)
^ |
'------------------ Feedback Loop -----------------'
| Layer | Purpose | Example |
|---|---|---|
| Requirement | What to build | "Users can log in with email" |
| Specification | How to build it | "OAuth2 + JWT, sessions stored in Redis" |
| GitHub Issue | Concrete implementation tasks | "Implement POST /auth/login endpoint" |
Each layer links to the others. When a requirement changes, you can trace the impact through specifications to issues. When implementation feedback surfaces, it flows back to update requirements.
Lifecycle:
draft --> pending_approval --> approved --> implemented --> deprecated
Requirements move through a defined lifecycle with PR-based approval gates, so nothing gets lost and nothing ships without review.
npm install -g @reqord/cli
# Optional: install the web dashboard
npm install -g @reqord/web
Reqord provides a Claude Code plugin for AI-assisted requirements workflow (design, TDD implementation, review, Git operations).
# Add the marketplace
/plugin marketplace add kicchann/reqord
# Install the plugin
/plugin install reqord@reqord-plugins
cd /path/to/your/project
# Initialize the .reqord/ directory structure
reqord init
# Set up project context
reqord context init
# Create a requirement (supports user-story, ears, and free-form formats)
reqord req create
# List all requirements
reqord req list
# Validate requirement quality with SMART scoring
reqord req validate req-000001
See the Getting Started guide for the full walkthrough.
Prerequisites: Node.js 20+, pnpm 10+, Git
git clone https://github.com/kicchann/reqord.git
cd reqord
pnpm install
pnpm build
cd packages/cli && pnpm link --global
Requirements are stored as YAML metadata (status, priority, dependencies, version history) paired with Markdown content (descriptions, success criteria, use cases). Machine-readable structure with human-readable documentation.
Built-in quality scoring based on the SMART framework (Specific, Measurable, Achievable, Relevant, Time-bound). Run reqord req validate to get an objective quality score and actionable improvement suggestions for any requirement.
npx claudepluginhub kicchann/reqord --plugin reqordRequirements-driven development workflow with quality gates for practical feature implementation
AI assistant framework for sphinx-needs projects: change analysis, traceability, MECE, authoring, verification, and release management
Comprehensive AI-assisted requirements elicitation. Supports stakeholder interviews (LLMREI pattern), document extraction, stakeholder simulation, domain research, gap analysis, user story mapping, customer journey mapping, JTBD analysis, prioritization (MoSCoW/Kano/WSJF), surveys, workshops, brainstorming, and business rules analysis. Exports to canonical, EARS, and Gherkin formats.
Specification-driven development workflow: specify → plan → tasks → implement
Spec-driven development for big features. When features get too big, plan mode gets too vague—leading to hallucinations during implementation. ShipSpec replaces vague plans with structured PRDs, technical designs, and ordered tasks that keep Claude grounded.
GitHub Spec-Kit integration for Specification-Driven Development - define WHAT and HOW before coding