By morodomi
Follow a structured 7-phase TDD workflow (INIT→PLAN→RED→GREEN→REFACTOR→REVIEW→COMMIT) with AI agents that review code designs for architectural consistency, risks to existing features, and scoped changes (≤10 files). Auto-generate conventional Git commits post-review, updating STATUS.md and running pre-commit hooks.
PLANフェーズの設計を担当するエージェント。Cycle docを受け取り、Skill(tdd-plan)を実行して設計・Test Listを作成する。
設計整合性レビュー。アーキテクチャ、パターン、レイヤー構造をチェック。
コード正確性レビュー。論理エラー、エッジケース、例外処理をチェック。
GREENフェーズの実装を担当するワーカーエージェント。テストケースとCycle docを受け取り、最小限の実装を行う。
ガイドライン準拠レビュー。コーディング規約、命名規則、ドキュメントをチェック。
PLANフェーズの設計を5観点で並行レビュー。信頼スコア80以上でBLOCK。
コード変更を6観点で並行レビュー。信頼スコア80以上でBLOCK。tdd-reviewで自動実行。
変更をGitコミットしてTDDサイクルを完了する。REVIEWの次フェーズ。「コミットして」「commit」で起動。
複雑なバグの並列仮説調査。Subagent/Agent Teams両モード対応。「原因調査」「investigate」「diagnose」で起動。
テストを通すための最小限の実装を行う。REDの次フェーズ。「実装して」「green」で起動。
Uses power tools
Uses Bash, Write, or Edit tools
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.
Claude Code plugin for enforcing strict TDD workflows
v5.1.1: Agent Memory - Reviewers and advisors accumulate project-specific knowledge across sessions
When letting AI write code, these problems often occur:
tdd-skills solves these by "enforcing" TDD on Claude Code.
claude
> /plugin marketplace add morodomi/tdd-skills
> /plugin install tdd-core@tdd-skills
> /plugin install tdd-php@tdd-skills # Choose based on language
> TDD setup
# or "onboard"
# Automatically executes:
# - Framework detection
# - CLAUDE.md generation
# - docs/ structure creation
> I want to add a login feature
# TDD cycle starts automatically:
# INIT -> PLAN -> RED -> GREEN -> REFACTOR -> REVIEW -> COMMIT
# Register marketplace
/plugin marketplace add morodomi/tdd-skills
# Install TDD workflow
/plugin install tdd-core@tdd-skills
# Language-specific quality tools (choose based on project)
/plugin install tdd-php@tdd-skills # PHP
/plugin install tdd-laravel@tdd-skills # Laravel
/plugin install tdd-wordpress@tdd-skills # WordPress / Bedrock
/plugin install tdd-python@tdd-skills # Python
/plugin install tdd-flask@tdd-skills # Flask
/plugin install tdd-ts@tdd-skills # TypeScript
/plugin install tdd-js@tdd-skills # JavaScript
/plugin install tdd-hugo@tdd-skills # Hugo SSG
/plugin install tdd-flutter@tdd-skills # Flutter / Dart
# Multi-language projects (e.g., Laravel + Alpine.js)
/plugin install tdd-php@tdd-skills
/plugin install tdd-js@tdd-skills
INIT -> PLAN -> RED -> GREEN -> REFACTOR -> REVIEW -> COMMIT
| Phase | Skill | Description |
|---|---|---|
| INIT | tdd-init | Create cycle document, scope confirmation |
| PLAN | tdd-plan | Design & planning |
| RED | tdd-red | Create failing tests (parallel) |
| GREEN | tdd-green | Minimal implementation (parallel) |
| REFACTOR | tdd-refactor | Code improvement |
| REVIEW | tdd-review | Quality check |
| COMMIT | tdd-commit | Git commit |
When CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS is enabled:
| Feature | Without Agent Teams | With Agent Teams |
|---|---|---|
| quality-gate | 6-agent parallel (subagent) | Debate mode (discuss & refute) |
| tdd-diagnose | Explore agent parallel investigation | Team debate investigation |
| tdd-parallel | Not available (sequential fallback) | Cross-layer parallel development |
Bug report -> Hypothesis generation -> Parallel investigation -> Root cause
|
Investigator 1: "Race condition in auth"
Investigator 2: "Cache invalidation issue"
Investigator 3: "DB connection timeout"
|
Debate & refute -> Root cause identified
INIT -> PLAN -> [tdd-parallel] -> REVIEW -> COMMIT
|
Teammate A: Backend (RED->GREEN->REFACTOR)
Teammate B: Frontend (RED->GREEN->REFACTOR)
Teammate C: Database (RED->GREEN->REFACTOR)
|
Integration test -> All green
Claude acts as a PdM (Product Manager), focusing on planning and decision-making while delegating implementation, testing, and review to specialist agents. This separates concerns and keeps the main context lightweight.
v4.3: Claude --- Read Skill --- Execute phases --- Some subagent parallelization
v5.0: Claude(PdM) --- Plan & Decide --- Delegate --- Autonomous judgment
|
+-----+-----+
architect red/green refactorer
| Block | Phases | Description |
|---|---|---|
| Planning | INIT -> PLAN -> plan-review | PdM plans, architect agent designs |
| Implementation | RED -> GREEN -> REFACTOR -> quality-gate | Worker agents implement & review |
| Finalization | COMMIT | PdM commits with full context |
| Agent | Phase | Role |
|---|---|---|
| architect | PLAN | Design and test list creation via Skill(tdd-plan) |
| red-worker | RED | Test creation (parallel) |
| green-worker | GREEN | Minimal implementation (parallel) |
| refactorer | REFACTOR | Code quality improvement via Skill(tdd-refactor) |
Flutter/Dart quality tools: dart analyze, dart format, flutter test
TypeScript quality tools: tsc, ESLint, Prettier, Jest/Vitest
WordPress-specific quality tools for TDD workflow. PHPStan for WordPress, WPCS, WP_UnitTestCase patterns.
Flask quality tools for TDD workflow. pytest-flask, mypy, Black/isort integration.
Laravel-specific quality tools for TDD workflow. Pest Laravel Plugin, HTTP tests, Eloquent patterns.
npx claudepluginhub morodomi/tdd-skills --plugin tdd-coreComprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
Lazy senior dev mode. Forces the simplest, shortest solution that actually works: YAGNI, stdlib first, no unrequested abstractions.
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
Develop, test, build, and deploy Godot 4.x games with Claude Code. Includes GdUnit4 testing, web/desktop exports, CI/CD pipelines, and deployment to Vercel/GitHub Pages/itch.io.
A growing collection of Claude-compatible academic workflow bundles. Covers scientific figures, manuscript writing and polishing, reviewer assessment, citation retrieval, data availability, paper reading, literature search, response letters, paper-to-PPTX conversion, and evidence-grounded Chinese invention patent drafting. Rules are organized as reusable skill folders with explicit workflows and quality checks.