convez.ai
Skill
This-HW

claude-code-kit

This-HW/claude-code-kit

Ver no GitHub
InicialPluginScore · de 0 a 100

Claude Code Plugin — 52 agents + 19 skills for everyone

Instalação

$npx convez-ai This-HW/claude-code-kit

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
4
Forks
0
Acompanhando
0
Saúde
42%
Último commit
há 2 meses
Versão
v2.10.4
Issues abertas
0
Contribuidores
1

README

A tradução para o português deste README ainda está na fila.

claude-code-kit

Universal Claude Code toolkit — 33 agents + 14 skills for software development

A focused, single-plugin AI agent system built for Claude Code. Covers the full software development lifecycle: planning, implementation, review, testing, and meta-tooling.

A single, well-tested core plugin built on a native-first foundation, scale-appropriate orchestration, a feedback learning loop, loop engineering, and a Definition-of-Done gate. (v2.7.0 · docs/specs/)


Quick Install

Prerequisites: Claude Code CLI installed (claude --version)

# Add the marketplace and install (recommended — works today)
/plugin marketplace add This-HW/claude-code-kit
/plugin install claude-code-kit@This-HW/claude-code-kit

# Updating: refresh the marketplace, then the new version is picked up
/plugin marketplace update claude-code-kit

Official registry: submission to claude-plugins-official is pending Anthropic review. Once approved, /plugin install claude-code-kit@claude-plugins-official will also work. Until then, use the marketplace command above.

Full Mode (Security Hooks + Auto-format)

git clone https://github.com/This-HW/claude-code-kit
cd claude-code-kit
./setup.sh

Options:

  • ./setup.sh --status — Check setup state
  • ./setup.sh --migrate — Migrate from legacy .claude/agents/ setup
  • ./setup.sh --force — Reset and re-run setup

Architecture & Concepts

claude-code-kit이 무엇을 어떻게 융합하는지 — 한눈에 보는 설계 원리.

설계 철학 — 네이티브 우선, kit은 의견 레이어

기술부채의 최대 원천은 Claude Code가 네이티브로 하는 일을 자체 구현으로 중복하는 것이다.

네이티브 프리미티브(agents·skills·hooks·dynamic workflows·OTEL·memory)는 거의 매주 진화한다. 그래서 이 kit의 가치는 인프라가 아니라, 그 위에 얹는 "의견이 담긴 에이전트·스킬·규율 레이어" 다. 관측·위임·훅 실행 같은 인프라는 네이티브에 위임하고, 손으로 만든 대체물은 삭제한다 → zero-debt.

네 갈래의 융합

갈래무엇을 가져왔나kit에서
Claude Code 네이티브agents, skills, hooks, dynamic workflow, OTEL, memory토대 프리미티브 — 매니페스트 의존성, exec-form 훅, decision:block 자동수정, 네이티브 관측
superpowers 규율brainstorming→plan→execute, phase gate, TDD, verification-before-completionbrainstorming → plan-task → auto-dev HARD-GATE 체인, Iron Law 검증
Hermes 피드백 루프"메모리·피드백 루프가 코어"validation 결함 → feedback ledger → 다음 구현 컨텍스트 주입 (학습 루프)
자체 Work 시스템파일 기반 감사 가능 추적docs/works/ Work ID·progress.md·decisions.md

Harness × Loop Engineering

두 상보 개념이 kit의 자율성을 만든다:

  • Harness Engineering — 어디서·무엇으로 행동하는가. 컨텍스트 주입(session-start), 도구 큐레이션(per-agent tools), 가드레일(protect-sensitive·stop-validator), Work 메모리.
  • Loop Engineering — 얼마나 오래·끈질기게 행동하는가. 승인된 배치를 P0·완료·가드 전까지 자율 완주. 게이트(설계·사람 멈춤)와 루프(실행·자율)를 분리한다.

결합 방식

 [설계 게이트 — 사람 승인]              [실행 루프 — 자율 완주]
 brainstorming → plan-task    ──승인──▶  auto-dev 배치 드라이버
 (무엇을 만들지 HARD-GATE)               (TaskList 폴링 + 종료 가드)
                                              │
        ┌─────────────────────────────────────┤ 스케일별 오케스트레이션
        ▼                                     ▼
   Small/Medium: 스킬 주도 플랫 dispatch   Large: 네이티브 ultracode
        │                                     │
        ▼  validation (review + security)     ▼
   continueOnBlock 자동수정 마이크로루프
        │
        ▼  결함 → feedback ledger → 다음 세션 LESSONS 주입  ◀─┐
        └──────────────────── 학습 루프 ──────────────────────┘

세부는 SSOT 문서 참조: CLAUDE.md (오케스트레이션·규율), docs/specs/ (설계 스펙), plugins/common/rules/ (규칙).


핵심 개념 & AI 엔지니어링 로직

kit에 녹아 있는 개념과 그 장점 — 어떻게 구현되는지와 함께.

개념 / 로직kit에서 어떻게장점
Native-first (zero-debt)네이티브 프리미티브를 최대 활용하고 자체 중복 구현은 삭제유지보수 부채 0, 네이티브가 진화해도 항상 최신
Phase-gate disciplinebrainstorming → plan-task → auto-dev HARD-GATE 체인모호성 100% 제거 후 구현 → 재작업·헛수고 최소화
Verification-before-completion (DoD)Iron Law + scripts/verify-done.sh 기계 게이트증거 없는 "완료" 주장을 구조적으로 차단
Loop engineering게이트(사람 멈춤) vs 루프(자율 완주) 분리 + 배치 드라이버 + 종료 가드P0 전까지 자율 완주, 런어웨이 방지
Feedback learning loop (Hermes)validation 결함 → ledger(상한·중복제거·감쇠) → 다음 세션 === LESSONS === 주입같은 실수를 반복하지 않음
Scale-appropriate orchestrationSmall/Medium 스킬 주도 플랫, Large 네이티브 ultracode 위임스케일별 최적, main 컨텍스트 병목 회피
Adversarial reviewreview-code가 4 페르소나(hacker·murphy·future-self·picky-user)로 침투 검토버그·엣지케이스를 능동 발굴
Multi-perspective deliberation10 관점 × 3 라운드 합의(/multi-perspective-review) + devil's advocate설계 사각지대 제거
Agent specialization33 전문 에이전트 × 모델 티어(Opus 전략 / Sonnet 구현 / Haiku 탐색)작업별 최적 모델·비용
Worktree isolation파일 수정 에이전트를 격리 git worktree에서 실행병렬 작업 충돌 방지
Harness engineering컨텍스트 주입(session-start)·도구 큐레이션·가드레일 훅·Work 메모리환경이 모델을 올바른 궤도로 유지
SSOT governancerules/ + decisions 추적 + 거버넌스/시크릿 보호 훅일관성·감사 가능성

Works with superpowers

obra/superpowers 플러그인과 상호보완하도록 설계됐습니다 — 둘을 같이 켜도 충돌·중복이 없습니다.

  • 각자 자동 적용: 둘 다 세션 시작에 자기 메타스킬을 자동 주입 (using-claude-code-kit / using-superpowers). 수동 호출 불필요.
  • 중복 제거: using-claude-code-kit은 범용 스킬 규율(1% 룰·red flags)을 superpowers에 양보하고, kit 고유 델타(에이전트맵·Work 시스템·native/loop/DoD)만 제공 → 병행 시 중복 0.
  • 역할 분담: superpowers = 방법론 지휘자, claude-code-kit = 실행 레이어(전문 에이전트·auto-dev 파이프라인·hooks·Work 추적).
  • 시너지: kit의 Definition-of-Done(기계 게이트) + superpowers의 verification-before-completion(원칙)이 상호보강.
  • 단독 동작: superpowers 없이 claude-code-kit만으로도 자급자족.

What's Included

PluginAgentsSkillsDescription
claude-code-kit3314Core: planning, development, review, meta

Architecture

2-Tier Agent Model

Tier 1: plugins/common/  — Core agents for all projects (33 agents)
Tier 2: project-local/   — Project-specific agents (user-defined)

Phase Gate Pattern

All workflows follow a 3-phase gate:

Phase 1 (Planning)     → Remove 100% ambiguity via planning agents
Phase 2 (Development)  → Implement based on Phase 1 artifacts
Phase 3 (Validation)   → Parallel review + security scan

Delegation Signal

Every agent ends its response with a structured handoff:

---DELEGATION_SIGNAL---
TYPE: DELEGATE_TO | TASK_COMPLETE | NEED_USER_INPUT
TARGET: [agent-name]
REASON: [reason]
CONTEXT: [handoff context]
---END_SIGNAL---

Model Selection

ModelUse CaseExamples
OpusStrategy, analysis, reviewclarify-requirements, review-code, plan-implementation
SonnetCode implementation, fixesimplement-code, fix-bugs, write-tests
HaikuExploration, quick checksexplore-codebase, verify-code, enforce-structure

Worktree Isolation

File-modifying agents run in an isolated git worktree to prevent conflicts:

  • implement-code, fix-bugs, write-tests
  • write-api-tests, write-ui-tests

Components (Core)

Key Skills

SkillCommandDescription
plan-task/plan-task5-phase planning pipeline: explore → clarify → journey → logic → implement
auto-dev/auto-devFull automated development pipeline
web-research/web-researchMCP-powered research: Context7 (docs) + Exa (code) + Tavily (web)
review/reviewCode review pipeline: ruff + review-code + security-scan
multi-perspective-review/multi-perspective-review3-Round Deliberation: 10 perspectives, consensus-driven
doc-coauthoring/doc-coauthoringAI-assisted documentation authoring and review
debug/debug4-Phase debug: diagnose → fix-bugs → verify-code
test/testRun tests and auto-fix failures via verify-code + fix-bugs
agent-creator/agent-creatorGenerate claude-code-kit plugin agents with correct frontmatter
skill-creator/skill-creatorGenerate claude-code-kit skills with best practices
mcp-builder/mcp-builderScaffold MCP servers and configure Claude Code integration
agent-teams/agent-teamsLarge-scale parallel work — routes to native ultracode (dynamic workflow)

Planning Agents (5 — Opus)

Read-only. No file modifications. Used in Phase 1.

AgentDescription
clarify-requirementsDetects ambiguous requests, generates P0/P1/P2 questions
analyze-domainDDD-based domain analysis, bounded context identification
define-business-logicDefines policies, rules, calculations, state transitions (CALC/VAL/STATE/POL)
design-user-journeyUX flows, screen design, onboarding, payment processes
define-metricsKPI, SLO, SLA, dashboard metric definitions

Meta Agents (6 — Opus)

Orchestrate multi-perspective review workflows. No Bash access.

AgentDescription
facilitatorAnalyzes what perspectives are needed, assigns agents
facilitator-teamsManages Round 0/2/3, integrates facilitator + synthesizer + consensus-builder
synthesizerConsolidates Round 1/2 results, identifies conflicts and duplicates
devils-advocateFailure scenario analysis via 4 attack personas (scalability / dependency / maintainability / cost)
consensus-builderConflict analysis across perspectives → Win-Win resolution
impact-analyzerSystem-wide impact, risk, and development cost of proposed changes

Backend Agents (4)

AgentModelDescription
design-servicesOpusClean/Hexagonal architecture, microservices, DDD patterns
implement-apiSonnetREST/GraphQL API implementation (Express, FastAPI, NestJS)
write-api-testsSonnetAPI unit / integration / E2E tests
optimize-logicSonnetAlgorithm optimization, caching, N+1 query fixes

Dev Agents (18)

Core development workflow agents.

AgentModelDescription
explore-codebaseHaikuProject structure, dependencies, pattern analysis
plan-implementationOpusRequirements → tech decisions → task breakdown → risk analysis
implement-codeSonnetCode implementation (worktree isolated)
write-testsSonnetUnit / integration / E2E tests (worktree isolated)
review-codeOpusAdversarial review via 4 personas: hacker, murphy, future-self, picky-user
fix-bugsSonnetMinimal-change bug fixes (worktree isolated)
verify-codeHaikuType check, lint, build, test execution
security-scanSonnetOWASP Top 10, secret exposure, vulnerable component detection
verify-integrationHaikuConnection integrity, data flow, version compatibility
git-workflowSonnetBranches, PRs, commit messages, merge strategies
sync-docsSonnetAPI and architecture documentation sync
plan-refactorOpusStructural improvement planning, ARCHITECTURE_LIMIT resolution
analyze-dependenciesSonnetLibrary versions, security updates
manage-api-versionsOpusAPI versioning strategy, migration, backwards compatibility
analyze-tech-debtSonnetCode quality analysis, tech debt prioritization
research-externalSonnetExternal library / technology / best practice research
generate-boilerplateSonnetProject templates, base structure generation
enforce-structureHaikuFile placement, naming convention compliance

Typical Workflows

Feature Development

clarify-requirements → analyze-domain → design-user-journey → define-business-logic
  → plan-implementation → implement-code → write-tests → verify-code
  → review-code + security-scan (parallel) → fix-bugs → sync-docs

Multi-perspective Review

/multi-perspective-review
  → facilitator (assigns perspectives)
  → [devils-advocate + synthesizer + impact-analyzer] (Round 1 parallel)
  → synthesizer (Round 2 consolidation)
  → consensus-builder (Round 3 resolution)

Debug

/debug
  → diagnose → fix-bugs → verify-code → (loop until green)

Security

  • Hooks: protect-sensitive.py runs on every Edit — blocks commits containing secrets
  • Auto-format: auto-format.py runs after edits (uses ruff for Python)
  • CI: gitleaks scans all pushes to main/stable
  • Policy: Never hardcode API keys, secrets, or internal IPs

Project Structure

plugins/
└── common/      — Core agents (33) + skills (14) + rules (12) + hooks

The plugin contains:

  • .claude-plugin/plugin.json — plugin manifest
  • agents/ — agent .md files with YAML frontmatter
  • skills/ — skill .md files
  • hooks/ — Python hook scripts
  • rules/ — governance rules

Contributing

PRs welcome. Checklist:

  • Agent frontmatter has name, description, model, maxTurns
  • Description includes MUST USE when: trigger conditions
  • No forbidden fields: permissionMode, context_cache, output_schema, next_agents, inline hooks
  • File-modifying agents have isolation: worktree
  • Regular agents have disallowedTools: [Task]
  • Skill description field is in English
  • Registered in plugin.json (with homepage, repository, license, author.email)
  • CI passes (JSON valid, frontmatter complete, no forbidden fields, pytest green, no secrets)

License

MIT

Ver o original no GitHub

Bom para

Relacionados