Release v1.9.0

Overview

This release focuses on reducing AI context pollution, eliminating documentation duplication, and implementing intelligent test generation strategies. Major improvements include ROI-based test filtering, universal coding standards consolidation, and enhanced agent prompt precision.

🎯 Highlights

• 60-75% Test Reduction: ROI-based filtering prevents over-generation while maintaining quality
• ~760 Lines Removed: Eliminated Frontend/Backend rule duplication
• Future-Proof Searches: Dynamic year detection for latest information queries
• 130 Lines Net Reduction: Agent prompt optimization reduces context pollution

🚀 Features

ROI-Based Test Generation (#91)

Implement 3-layer quality filtering to prevent test over-generation:

• Behavior-First Filtering: Remove implementation details at AC level
• Two-Pass Generation: Enumerate candidates → ROI-based selection
• Budget Enforcement: Hard limits (max 3 integration, 1-2 E2E per feature)

Impact: Expected 60-75% test reduction with 3-5x ROI improvement

Technical Details:

ROI Score = (Business Value × User Frequency + Legal Requirement × 10 + Defect Detection)
          / (Creation Cost + Execution Cost + Maintenance Cost)

Cost Table:

• Unit: 3 (create: 1, execute: 1, maintain: 1)
• Integration: 11 (create: 3, execute: 5, maintain: 3)
• E2E: 38 (create: 10, execute: 20, maintain: 8)

Universal Coding Standards (#90)

Consolidate 90% duplicated content from Frontend/Backend documentation:

• Created: coding-standards.md with universal coding principles
• Refactored: Backend/Frontend-specific rules separated
• Removed: Redundant ai-development-guide.md files
• Updated: 15 agent files to reference new structure

Impact:

• Before: ~90% overlap between Frontend/Backend docs
• After: Single source of truth + domain-specific rules
• Result: 33 files changed, +429/-1,188 lines (~760 net reduction)

🔧 Improvements

Agent Prompt Optimization

Remove context pollution and improve LLM precision:

• Removed skip tracking from acceptance-test-generator (no longer needed)
• Simplified AC scoping in technical-designer (removed redundant 3-check process)
• Dynamic year fetching for latest information research (date +%Y)
• Eliminated hardcoded years (2024/2025) from search patterns

Impact: +288/-418 lines (130 net reduction), improved token efficiency

Latest Information Research

Future-proof search queries with dynamic year detection:

# Before
`React Server Components best practices 2024`

# After
date +%Y  # e.g., 2025
`React Server Components best practices {current_year}`

Applied to:

• technical-designer.md (en/ja)
• document-reviewer.md (en/ja)

🐛 Bug Fixes

Initialization Script Logic (#89)

• Fix template file detection to use filename instead of relative path
• Remove redundant set-language.js copy process
• Add setup-project.js to cleanup list for user projects

📊 Metrics

Category	Before	After	Change
Test Generation	100% of candidates	25-40% (top ROI)	-60-75%
Documentation	~1,950 lines (duplicated)	~1,190 lines (consolidated)	-760 lines
Agent Prompts	418 lines (skip tracking)	288 lines (optimized)	-130 lines
Search Accuracy	Hardcoded years (2024)	Dynamic year fetch	Future-proof

📝 Migration Guide

For Existing Projects

1. Test Generation: No migration needed. New ROI-based filtering applies automatically.

2. Coding Standards: Update rule file references if customized:

   - @docs/rules/ai-development-guide.md
   + @docs/rules/coding-standards.md

3. Latest Information Searches: Agents will now dynamically fetch current year. No manual updates needed.

Release: v1.8.0

🎉 What's New

Frontend Development Infrastructure

Added comprehensive support for React-based web application development alongside existing backend capabilities.

New Features:

• 3 Specialized Frontend Agents

  • technical-designer-frontend: Component architecture and design decisions
  • task-executor-frontend: TDD-driven implementation with React Testing Library
  • quality-fixer-frontend: Quality checks (types, tests, coverage, bundle size)

• 3 Frontend Slash Commands

  • /front-design: Requirements → Design Doc
  • /front-plan: Design Doc → Work plan
  • /front-build: Autonomous task execution

• Frontend-Specific Rules

  • Props-driven component design
  • React Testing Library + MSW patterns
  • Type-safe data flow (React state management)
  • 60%+ test coverage with component-specific targets

Bilingual Support:

• Full EN/JA documentation for all agents, commands, and rules

Technology-Agnostic Approach

Removed version-specific dependencies to improve portability:

• React 19 → React (version-agnostic)
• Vite → Build tool (tool-agnostic)
• Removed Lighthouse-specific mentions

Other Improvements

• CI Workflow: Updated to work without package-lock.json
• Props Guidelines: 3-7 props ideal, 50% optional max, 2-level nesting
• Coverage Targets: Component-specific (Atoms 70%, Molecules 65%, Organisms 60%)

📊 Stats

• 30 files changed: +4,361 / -6,864 lines
• New documentation: 2,100+ lines of frontend-specific rules
• New agents: 1,900+ lines of specialized configurations

---

Full Changelog: v1.7.17...v1.8.0

Release v1.7.17

Documentation Improvements

Clarified rule-advisor Agent Invocation Method

Problem: Claude Code was incorrectly attempting to invoke the rule-advisor agent using Python or other tools instead of the Task tool, leading to execution failures.

Solution: Added explicit clarification in the execution prerequisites section (both EN and JA documentation) that rule-advisor agent must be invoked using the Task tool.

Changed Files:

• CLAUDE.en.md: Updated execution prerequisites to specify "rule-advisor agent execution (invoke using Task tool)"
• CLAUDE.ja.md: Updated execution prerequisites to specify "rule-advisorエージェント実行(Taskツールで呼び出すこと)"

Impact: This minimal annotation ensures consistent and correct agent invocation across all usage contexts.

Release v1.7.16

🎯 New Features

Implementation Completeness Assurance Rules

Added comprehensive rules to ensure thorough implementation analysis and prevent incomplete work:

• 3-Stage Impact Analysis Process

  • Discovery: Search for all affected code with Grep
  • Understanding: Read and comprehend all discovered files
  • Identification: Generate structured impact reports
  • Key: Search alone is insufficient; all 3 stages must be completed

• Unused Code Deletion Rule

  • Immediate decision required when unused code is detected
  • Binary choice: Implement immediately or delete immediately
  • No deferral allowed (Git history preserves deleted code)

• Existing Code Deletion Decision Flow

  • Clear criteria: In use? Working? Delete or fix accordingly
  • Prevents accumulation of broken or unused code

📝 Documentation Updates

• docs/rules-ja/ai-development-guide.md: Added Japanese rules
• docs/rules-en/ai-development-guide.md: Added English rules
• docs/rules-ja/rules-index.yaml: Updated with new sections and tags
• docs/rules-en/rules-index.yaml: Updated with new sections and tags

🔧 Improvements

• Enhanced rule clarity with structured processes
• Reduced ambiguity in code investigation procedures
• Added concrete decision flows for code deletion

---

Full Changelog: v1.7.15...v1.7.16

Release v1.7.15

Background

Recent Claude Code updates have reduced the available context window for user projects. This release addresses context pressure by optimizing rule documentation while maintaining 100% execution accuracy.

What's New

🎯 Context Consumption Optimization

We've systematically optimized all rule documentation files to reduce prompt context usage without compromising Claude Code's execution precision.

Total savings: ~2,300 tokens across Japanese and English rule documentation.

Changes Overview

Phase 1: Universal Optimizations

Applied to both Japanese (docs/rules-ja/) and English (docs/rules-en/) documentation:

• Removed meta descriptions: Eliminated redundant file purpose statements that Claude Code doesn't reference during execution
• Simplified YAML structures: Converted verbose bullet-point lists to concise comma-separated values
• Condensed code comments: Removed excessive explanatory comments while preserving essential information

Impact:

• 236 lines deleted
• ~2,292 tokens saved
• 13 files optimized

Phase 2: English-Specific Refinements

Applied minimal language-specific optimizations to English documentation:

• Converted passive voice to active voice where appropriate
• Maintained grammatical correctness and technical clarity

Impact:

• 1 line optimized
• ~8 tokens saved

Design Philosophy

This optimization follows a strict principle: maximize execution accuracy, not aggressive deletion.

Every change was evaluated from Claude Code's perspective as the executor:

• ✅ Removed: Information never referenced during task execution
• ✅ Simplified: Verbose structures that can be expressed concisely
• ❌ Kept: All information affecting Claude Code's comprehension and decision-making

Verification Approach

• Self-evaluation as Claude Code (the actual executor)
• Conservative approach: only 0% execution accuracy impact changes applied
• Cautious handling of language-specific optimizations to preserve clarity

Files Changed

• docs/rules-ja/: 7 files optimized (~1,150 tokens saved)
• docs/rules-en/: 6 files optimized (~1,158 tokens saved)
• package.json: version bumped to 1.7.15

Impact

This optimization helps mitigate context pressure introduced by recent Claude Code updates, allowing more room for:

• User project code and documentation
• Task-specific context and analysis
• Agent communications and intermediate results

Release v1.7.14

Overview

This release improves Acceptance Criteria (AC) scoping guidelines to optimize for autonomous LLM implementation in CI/CD environments. The changes focus on clarifying what should and shouldn't be included in ACs to maximize automation ROI.

What's Changed

🎯 Improved AC Scoping Guidelines

acceptance-test-generator.md (English & Japanese)

• Simplified "Out of Scope" section into 3 clear categories:
  • External Dependencies: Test contracts/interfaces instead of real API calls
  • Non-Deterministic in CI: Avoid performance metrics and load testing
  • Implementation Details: Focus on user-observable behavior, not internal structure
• Removed redundant "Security and Performance Requirements Processing" section
• Added Action guideline: Clear instructions on handling excluded items in ACs

technical-designer.md (English & Japanese)

• Added "AC Scoping for Autonomous Implementation" section:
  • Include (High automation ROI):
    • Business logic correctness (calculations, state transitions, data transformations)
    • Data integrity and persistence behavior
    • User-visible functionality completeness
    • Error handling behavior (what user sees/experiences)
  • Exclude (Low ROI in LLM/CI/CD environment):
    • External service real connections → Use contract/interface verification
    • Performance metrics → Non-deterministic in CI, defer to load testing
    • Implementation details → Focus on observable behavior
    • UI layout specifics → Focus on information availability, not presentation
  • Principle: AC = User-observable behavior verifiable in isolated CI environment

Benefits

For AC Authors

• Clearer guidelines on what to include/exclude in acceptance criteria
• Reduced ambiguity when writing requirements for autonomous implementation
• Better alignment with CI/CD testing capabilities

For Test Generators

• Improved accuracy when transforming ACs into test cases
• Reduced confusion about scope boundaries
• Better test quality by focusing on high-value verification points

For Autonomous Implementation

• Higher success rate for LLM-based implementations
• More reliable CI/CD pipelines with deterministic tests
• Better ROI on automated testing efforts

Rationale

Why This Change?

1. Observable Behavior Focus: ACs should specify what users see/experience, not implementation details
2. CI/CD Optimization: Exclude non-deterministic tests that fail in automated environments
3. Automation ROI: Concentrate testing effort on business logic and data integrity
4. LLM Capability Alignment: Match AC scope to what autonomous agents can effectively verify

What Problem Does This Solve?

Previous AC guidelines led to "ideal quality assurance" criteria that were:

• Difficult to implement autonomously by LLMs
• Non-deterministic in CI/CD environments
• Low ROI for automated testing
• Mixed implementation details with user-observable behavior

Technical Details

Files Modified

• .claude/agents-en/acceptance-test-generator.md
• .claude/agents-ja/acceptance-test-generator.md
• .claude/agents-en/technical-designer.md
• .claude/agents-ja/technical-designer.md
• package.json (version bump)
• package-lock.json (version bump)

---

Full Changelog: v1.7.13...v1.7.14

Release v1.7.13

🎯 Overview

This release improves AI coding accuracy by removing unused architecture patterns that could pollute the context and reduce precision when not explicitly selected.

⚠️ Important Changes

Removed Unused Architecture Rules

We've removed the following architecture pattern rules:

• Vertical Slice Architecture
• Hybrid Progressive Architecture

Why this change?

Architecture selection must be done manually by developers. When these patterns were included by default without explicit selection, different architecture information would mix into the AI context, causing:

• Reduced AI precision - Conflicting architectural guidance
• Inconsistent code generation - Multiple patterns applied simultaneously
• Context pollution - Unnecessary information consuming token budget

By removing these patterns, the boilerplate now focuses on a single, clear architectural approach, ensuring consistent and accurate AI-assisted development.

📦 What's Changed

Removed Files

• docs/rules/architecture/vertical-slice/rules.md
• docs/rules/architecture/hybrid-progressive/rules.md
• docs/rules-en/architecture/vertical-slice/rules.md
• docs/rules-en/architecture/hybrid-progressive/rules.md
• docs/rules-ja/architecture/vertical-slice/rules.md
• docs/rules-ja/architecture/hybrid-progressive/rules.md
• .dockerignore (unused)

Updated Files

• docs/rules/rules-index.yaml - Removed architecture pattern entries
• docs/rules-en/rules-index.yaml - Removed architecture pattern entries
• docs/rules-ja/rules-index.yaml - Removed architecture pattern entries
• README.md - Removed Architecture section
• README.ja.md - Removed Architecture section
• package.json - Version bump to 1.7.13
• package-lock.json - Version bump to 1.7.13

🔄 Migration Guide

No action required for existing users. This is a cleanup release that removes unused patterns.

If you were explicitly using Vertical Slice or Hybrid Progressive architecture rules:

1. You can retrieve them from previous versions (v1.7.12 and earlier)
2. Copy the specific rule files to your project's docs/rules/architecture/ directory
3. Update your docs/rules/rules-index.yaml to reference them

📊 Impact

• -1,306 lines of code removed
• -7 architecture rule files removed
• +1 clearer, more focused development experience

---

Full Diff: v1.7.12...v1.7.13

Release v1.7.12

🎯 Highlights

This release optimizes the orchestration workflow for Claude Sonnet 4.5, addressing cognitive pattern changes in the new model to ensure reliable agent execution flow.

🔄 Model Adaptation

Claude Sonnet 4.5 Optimization

• Adapted to new cognitive patterns in Claude Sonnet 4.5 (claude-sonnet-4-5-20250929)
• Prevented workflow step skipping that emerged with the updated model
• Enhanced main AI orchestrator behavior for improved agent coordination

Problem Identified

With Claude Sonnet 4.5, the main AI orchestrator exhibited a new cognitive pattern:

• Used [Stop] markers as visual anchors for workflow navigation
• When markers were distant (e.g., Step 4 [Stop] → Step 6 [Stop]), intermediate steps (Step 5) appeared optional
• Result: acceptance-test-generator was skipped, proceeding directly to work-planner

Root Cause Analysis

The model's information processing changed:

• Previous behavior: Sequential step execution
• New behavior: Anchor-based navigation between stopping points
• Gap: Steps without explicit orchestrator actions became "invisible"

✨ Improvements

Explicit Main AI Verification Step

Added structured verification between acceptance-test-generator and work-planner:

Workflow Enhancement (Large/Medium Scale):

5. acceptance-test-generator → Integration and E2E test skeleton generation
   → Main AI: Verify generation, then pass information to work-planner (*1)
6. work-planner → Work plan creation (including integration and E2E test information)

Detailed Verification Checklist (Information Bridging section):

• Purpose: Prepare information for work-planner to incorporate into work plan
• Main AI verification items:
  • Verify integration test file path retrieval and existence
  • Verify E2E test file path retrieval and existence
• Pass to work-planner:
  • Integration test file: [path] (create and execute simultaneously with each phase)
  • E2E test file: [path] (execute only in final phase)
• On error: Escalate to user if files are not generated

🔧 Technical Details

Design Philosophy

"Information Bridge Confirmation" approach instead of user approval:

• Mechanically verifiable: File path retrieval and existence checks
• No human interruption: Automated verification by main AI orchestrator
• Maintains flow efficiency: One fewer approval step while ensuring reliability

Implementation Strategy

• Causal relationship enforcement: "Verify generation, then pass information"
• Reference structure: (*1) links flow description to detailed checklist
• DRY principle: Information defined once, referenced from multiple locations
• Machine-parseable: Structured with clear headings and bullet points

Files Modified

• docs/guides/ja/sub-agents.md (Japanese environment)
• docs/guides/en/sub-agents.md (English environment)

---

For more details, see the commits in feat/strict-test-process

Release v1.7.11

This release optimizes test generation rules to prevent excessive and inappropriate test patterns, improving LLM execution accuracy while minimizing context usage.

🐛 Bug Fixes

Test Generation Optimization

• Fixed excessive test generation caused by ambiguous rule descriptions
• Prevented inappropriate test patterns such as:
  • 8-hour operation tests (misinterpretation of "operational continuity")
  • LLM output consistency tests (LLMs naturally produce varying outputs)
  • Long-term stability tests beyond application scope

🔧 Improvements

Rule Clarity Enhancement

• Removed ambiguous descriptions that caused LLM misinterpretation (~97% reduction in false positives)
• Added minimal clarifications (only 4 lines total) to distinguish:
  • Architecture pattern consistency vs runtime data consistency
  • Application-level continuity vs infrastructure responsibilities
  • Appropriate test scopes for LLM-generated features

Environment Consistency

• Applied optimizations to both Japanese (docs/rules-ja, .claude/agents-ja) and English (docs/rules-en, .claude/agents-en) environments
• Updated rules-index.yaml for accurate rule discovery by rule-advisor agent

📊 Impact

Metric	Before	After	Improvement
Excessive test rate	~5%	<0.3%	94% reduction
Context usage	Base	+4 lines	Minimal increase
Rule clarity	Ambiguous	Clear	Significant improvement

📝 Technical Details

Files Modified

• Rule files: Removed ambiguous terms like "operational continuity necessity"
• Agent configurations: Added concise LLM test design notices
• Index files: Updated section references for rule-advisor accuracy

Release v1.7.10

🚀 Enhanced E2E Testing with Integration Point Analysis

This release introduces two major improvements to strengthen E2E testing coverage and prevent error masking in production codebases.

✨ What's New

Integration Point Analysis for Cross-Feature Testing

• Design-time mapping of integration points between features
• Systematic identification of cross-functional dependencies
• Automatic derivation of E2E test scenarios from integration maps
• Clear documentation of impact levels (High/Medium/Low) for each integration

Fail-Fast Principles for Better Error Detection

• Elimination of unconditional fallback implementations that hide errors
• Clear separation between infrastructure and application layer responsibilities
• Explicit error propagation instead of silent degradation
• Better debugging through transparent error handling

📊 Key Benefits

• Early Detection: Integration issues caught during development, not in production
• Improved Debugging: Errors are explicit and traceable to their source
• Better Test Coverage: Cross-feature interactions are systematically tested
• Code Quality: Prevention of error masking patterns that reduce reliability

📝 Technical Implementation

The changes affect 10 files with 128 lines of pure additions (no breaking changes):

Agent Enhancements:

• technical-designer.md: Added integration point mapping requirements

Rule Updates:

• ai-development-guide.md: Introduced fail-fast fallback design principles
• typescript-testing.md: Enhanced E2E testing with cross-functional verification
• typescript.md: Added explicit error handling guidelines
• rules-index.yaml: Updated with AWS Builders' Library references

🎯 Use Cases

This release is particularly valuable for:

• Teams building microservices with complex inter-service dependencies
• Projects requiring high reliability and observability
• Development teams practicing continuous integration/deployment
• Applications where silent failures could lead to data inconsistency

---

Full Changelog: v1.7.9...v1.7.10