Orchestrate evidence-driven deep work sessions enforcing a 5-phase protocol: brainstorm goals, research codebases, plan TDD slices with git worktrees, implement via strict RED-GREEN-REFACTOR cycles, test with quality gates, drift detection, and receipt validation. Resume, fork, cleanup sessions; generate compliance reports and history stats.
npx claudepluginhub sungmin-cho/claude-deep-suite --plugin deep-workAssumption health report — validate whether deep-work's enforcement rules are still justified by session evidence
Phase 0: 문제 정의 및 접근법 탐색 — 왜(why)를 먼저 탐구.
Clean up stale deep-work worktrees — scan and delete old worktrees
Systematic debugging sub-mode — root cause investigation before fixes
Finish a deep work session — merge, PR, keep, or discard the branch
Fork a deep-work session to explore a different approach
View deep work session history — cross-session trends, model usage, TDD compliance
Phase 3: TDD 기반 슬라이스 구현 — RED → GREEN → REFACTOR.
코드 메트릭, 복잡도, 의존성 분석 기반 인사이트 리포트. 독립 실행 또는 deep-work Insight Quality Gate로 사용.
Run mutation testing on changed files to verify AI-generated test quality. Automatically fixes surviving mutants.
Manually trigger structural review and/or cross-model adversarial review on current phase document
Phase 2: 구현 계획 수립 — 슬라이스 분해, 의존성 매핑.
Receipt management — view, dashboard, export receipts
Generate or view the deep work session report
Phase 1: 코드베이스 심층 분석 — Solo/Team 자동 분기.
Resume an active deep work session — restores context and continues from where you left off
SOLID design principles code review — evaluates SRP, OCP, LSP, ISP, DIP compliance. Standalone or deep-work Advisory Quality Gate.
Run computational sensors (linter, type checker, coverage) on the project. Works independently of deep-work sessions.
Slice management — view status, activate, deactivate slices
Check the current deep work session status and progress
Phase 4: 종합 검증 — 센서, 드리프트, 커버리지.
Evidence-Driven Development Protocol — Brainstorm → Research → Plan → Implement → Test 자동 진행.
Plan alignment verification — detects unimplemented items, out-of-scope changes, and design drift. Standalone or deep-work Required Gate.
Phase 0 — Brainstorm: explore why before how (skip-able)
Phase 3 — Implement: slice-based TDD execution of approved plan
Phase 2 — Plan: create a detailed implementation plan
Phase 1 — Research: exhaustively analyze the codebase
Phase 4 — Test: comprehensive verification + implement-test retry loop
Evidence-Driven Development — session initialization + auto-flow orchestration
Evidence-driven development protocol with auto-flow orchestration. Use when: "deep work", "plan before code", "TDD", "evidence-driven", "분석 후 구현", "structured workflow", or complex multi-file tasks that benefit from structured planning before implementation.
Battle-tested Claude Code plugin for engineering teams — 38 agents, 156 skills, 72 legacy command shims, production-ready hooks, and selective install workflows evolved through continuous real-world use
Executes bash commands
Hook triggers when Bash tool is used
Modifies files
Hook triggers on file write and edit operations
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.
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
Claude Code skills for Godot 4.x game development - GDScript patterns, interactive MCP workflows, scene design, and shaders
Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer