refactor: Optimize AI agent definitions for improved execution accuracy #40
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
## 主な変更内容 ### エージェント定義ファイルの改善 - アーキテクチャルール参照を条件付き読み込みに変更（5ファイル） - 「入力情報」セクションを「必要情報」に簡潔化（6ファイル） - 「ドキュメント出力原則」を「出力方針」に統一・簡略化（3ファイル） ### 新規エージェント・コマンドの追加と改善 - code-reviewer.md: Design Doc準拠検証エージェントを追加 - 複雑度計測を実用的なチェック項目に簡略化 - 出力形式をJSON形式に変更し簡潔性を向上 - review.md: review-and-fix.mdをリネームし機能改善 - ユーザー確認フロー（y/n）を追加 - プロジェクト段階別の判定基準を導入 - メタ認知プロセスを明記 ### Design Docテンプレートの改善 - 受入条件セクションを追加し、テスト可能な条件定義を明確化 - 機能受入条件からテストケース導出の指針を追加 ### 不要なファイルの削除 - docs/rules-ja/architecture/hybrid-progressive/README.md - docs/rules-ja/architecture/vertical-slice/README.md （ボイラープレート利用者が配置することを想定） ## 改善効果 - コンテキスト理解の向上: 簡潔で明確な指示 - エラー防止: 存在しないファイル参照の条件付き化 - 実装可能性向上: 曖昧な指示を具体的な内容に変更 - 統一性向上: コマンド間で一貫した形式 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>
… improvements ## Changes ### English Agent Definitions Added/Updated - code-reviewer.md: New Design Doc compliance validation agent - Simplified complexity metrics to practical checklist items - Changed output format to concise JSON structure - task-executor.md, quality-fixer.md: Updated with conditional architecture rules - All agents: Standardized sections for better AI execution accuracy ### English Commands Added - review.md: Renamed from review-and-fix.md with improvements - Added user confirmation flow (y/n) before fixes - Introduced metacognition process for fix execution - Project stage-aware compliance criteria ### Consistency Improvements - Changed "Input Format/Information" → "Required Information" across all files - Unified "Document Output Principle" → "Output Policy" for brevity - Applied conditional loading for architecture rules (if present) ### Cleanup - Removed architecture rule README files (users can add project-specific rules) - docs/rules-en/architecture/hybrid-progressive/README.md - docs/rules-en/architecture/vertical-slice/README.md ## Translation Principles Applied - Prioritized clarity and actionability for AI execution - Maintained structural consistency with Japanese versions - Used imperative mood for instructions - Preserved metacognition markers and flow control elements 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>
## Changes ### Added to README.md - `/review` command: Design Doc compliance validation with optional auto-fixes - `/rule-maintenance` command: Development rule management - `code-reviewer` agent: Third-party Design Doc compliance validation ### Added to README.ja.md - `/review` コマンド: Design Doc準拠検証と自動修正 - `/rule-maintenance` コマンド: 開発ルール管理 - `code-reviewer` エージェント: 第三者視点での準拠検証 ## Details - Commands are properly placed in the existing "Claude Code Custom Slash Commands" section - Agents are added to the "Available subagents" section - Both English and Japanese versions updated consistently 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>
…h design template Aligned template-en.md with template-ja.md by adding: - Acceptance Criteria section for verifiable implementation conditions - Enhanced Testing Strategy with test design principles - E2E testing approach for complete scenario verification - Clear linkage between acceptance criteria and test cases These additions improve AI execution accuracy by providing clear, measurable success criteria. 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>
Minor version update for new features: - Code reviewer agent addition - Review command implementation - Enhanced agent definitions with improved execution accuracy 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR optimizes AI agent definitions to enhance execution accuracy and adds a new code-reviewer agent. Key improvements include:
#36