💻 코더

DDD/TDD 방법론으로 SPEC 요구사항 구현

문서, 코드맵 동기화 및 풀 리퀘스트 생성

코드베이스 스캔 및 AI 컨텍스트용 @MX 코드 수준 어노테이션 추가

EARS 형식 요구사항과 인수 기준을 포함한 SPEC 문서 생성

프로젝트 문서 생성 (product.md, structure.md, tech.md, codemaps/)

코드베이스 조사 서브에이전트 — 의존성·공개 API·영향 범위를 조사해 구조화 보고. claude-agentic-coding 워크플로우에서 위임. 사용자 프롬프트 불가 (blocker report 반환).

Unified artifact-meta creation specialist — builds the scaffolding/structure of agents, skills, plugins, commands, hooks, MCP servers, and LSP servers. Operates on artifact metadata (frontmatter, manifests, dispatch tables, hook registration) NOT artifact body content (prose, business logic, domain reasoning). Use PROACTIVELY for creating agents, skills, plugins, commands, hooks, MCP servers, and LSP servers. MUST INVOKE when ANY of these keywords appear in user request: EN: create agent, new agent, agent blueprint, sub-agent, agent definition, custom agent, create skill, new skill, skill optimization, knowledge domain, YAML frontmatter, create plugin, plugin, plugin validation, plugin structure, marketplace, new plugin, marketplace creation, marketplace.json, plugin distribution, create command, new command, create hook, MCP server, LSP server, lsp server KO: 에이전트생성, 새에이전트, 에이전트블루프린트, 서브에이전트, 에이전트정의, 커스텀에이전트, 스킬생성, 새스킬, 스킬최적화, 지식도메인, YAML프론트매터, 플러그인생성, 플러그인, 플러그인검증, 플러그인구조, 마켓플레이스, 새플러그인, MCP서버, LSP서버 JA: エージェント作成, 新エージェント, エージェントブループリント, サブエージェント, スキル作成, 新スキル, スキル最適化, YAMLフロントマター, プラグイン作成, プラグイン, プラグイン検証, マーケットプレイス, MCPサーバー, LSPサーバー ZH: 创建代理, 新代理, 代理蓝图, 子代理, 创建技能, 新技能, 技能优化, YAML前置信息, 创建插件, 插件, 插件验证, 市场, MCP服务器, LSP服务器 NOT for: SPEC body authoring (spec.md / plan.md / acceptance.md content — manager-spec only), code implementation, testing, documentation writing, git operations, production deployment

Unified implementation specialist (run-phase: implementation file authoring + owns progress.md §Run-phase Evidence/Audit-Ready Signal + draft → in-progress transition). See §SPEC Artifact Ownership for artifact-level boundaries. Supports three cycle_type modes: `tdd` (RED-GREEN-REFACTOR — default for new feature work), `ddd` (ANALYZE-PRESERVE-IMPROVE — legacy refactoring with characterization tests), and `autofix` (localize → repair → validate — invoked from the /moai fix pipeline workflow; routed via the `--mode` flag or pipeline class dispatch). Use PROACTIVELY for code implementation, refactoring, test-driven development, behavior preservation, and pipeline auto-fix execution. MUST INVOKE when ANY of these keywords appear in user request: EN (DDD): DDD, refactoring, legacy code, behavior preservation, characterization test, domain-driven refactoring EN (TDD): TDD, test-driven development, red-green-refactor, test-first, new feature, specification test, greenfield EN (autofix): autofix, auto-fix, /moai fix, lint repair, error repair, pipeline repair KO (DDD): DDD, 리팩토링, 레거시코드, 동작보존, 특성테스트, 도메인주도리팩토링 KO (TDD): TDD, 테스트주도개발, 레드그린리팩터, 테스트우선, 신규기능, 명세테스트, 그린필드 KO (autofix): 자동수정, 자동수리, 린트수정, 에러수정, 파이프라인수리 JA (DDD): DDD, リファクタリング, レガシーコード, 動作保存, 特性テスト, ドメイン駆動リファクタリング JA (TDD): TDD, テスト駆動開発, レッドグリーンリファクタ, テストファースト, 新機能, 仕様テスト, グリーンフィールド JA (autofix): 自動修正, 自動修復, リント修正, エラー修正, パイプライン修復 ZH (DDD): DDD, 重构, 遗留代码, 行为保存, 特性测试, 领域驱动重构 ZH (TDD): TDD, 测试驱动开发, 红绿重构, 测试优先, 新功能, 规格测试, 绿地项目 ZH (autofix): 自动修复, 自动修补, lint修复, 错误修复, 流水线修复 NOT for: SPEC body authoring (spec.md / plan.md / acceptance.md / design.md / research.md — manager-spec only per Status Transition Ownership Matrix), security audits, performance optimization, deployment (route domain-specialist work to a per-spawn Agent(general-purpose) per archived-agent-rejection.md §C)

Documentation specialist (sync-phase: CHANGELOG.md + README.md + docs-site authoring + owns progress.md §E.4 Sync-phase Audit-Ready Signal + the merged in-progress → implemented → completed transition on the single sync commit for all 4 SPEC artifacts, per the 3-phase close). See §SPEC Artifact Ownership for artifact-level boundaries — MUST NOT modify spec.md / plan.md / acceptance.md body content. Absorbs the project initialization and configuration role per the Anthropic catalog consolidation (17→8 agents; the prior project-doc-role owner is archived per .claude/rules/moai/workflow/archived-agent-rejection.md §C row 4) — product.md / structure.md / tech.md scaffolding and project-level documentation maintenance are now performed by this agent during /moai project and sync-phase. Use PROACTIVELY for README, API docs, Nextra, technical writing, markdown generation, and project documentation scaffolding. MUST INVOKE when ANY of these keywords appear in user request: EN: documentation, README, API docs, Nextra, markdown, technical writing, docs, project initialization, product.md, structure.md, tech.md KO: 문서, README, API문서, Nextra, 마크다운, 기술문서, 문서화, 프로젝트초기화, 제품문서, 구조문서, 기술문서 JA: ドキュメント, README, APIドキュメント, Nextra, マークダウン, 技術文書, プロジェクト初期化, プロダクト文書, 構造文書 ZH: 文档, README, API文档, Nextra, markdown, 技术写作, 项目初始化, 产品文档, 结构文档 NOT for: SPEC body authoring (spec.md / plan.md / acceptance.md body — manager-spec only per Status Transition Ownership Matrix; manager-docs limited to frontmatter `status` + `updated` field transitions only), code implementation, testing, git branch management, security audits

Git workflow specialist. Use PROACTIVELY for commits, branches, PR management, merges, releases, and version control. Invocation gate: invoked for Tier L SPEC PR creation OR explicit `--pr` flag per the canonical Tier-based PR routing policy. Tier S/M SPECs follow the Hybrid Trunk 1-person OSS pattern (main-direct push via manager-develop) per the Hybrid Trunk 1-person OSS policy; manager-git is NOT invoked for Tier S/M routine commits. MUST INVOKE when ANY of these keywords appear in user request: EN: git, commit, push, branch, PR, merge, release, rebase, tag, conventional commit KO: git, 커밋, 푸시, 브랜치, PR, 머지, 릴리즈, 리베이스, 태그 JA: git, コミット, プッシュ, ブランチ, PR, マージ, リリース, タグ ZH: git, 提交, 推送, 分支, PR, 合并, 发布, 标签 NOT for: Tier S/M default Hybrid Trunk main-direct (no PR step — handled by manager-develop), code implementation, testing, architecture design, documentation content, security audits

개발자용 Claude agentic 코딩 — tool use, sub-agents, MCP 연동으로 자율 코딩 워크플로우 구성. Claude 공식 tool-use/sub-agents 문서 기반. "Claude로 자율 코딩 에이전트 만들고 싶어" 질문에 즉시 활용.

Backend development specialist covering API design, database integration, microservices architecture, and modern backend patterns. Use when designing APIs, implementing server logic, authentication, or authorization.

Database specialist covering PostgreSQL, MongoDB, Redis, Oracle, and cloud database platforms (Neon, Supabase, Firestore). Use for schema design, query optimization, indexing strategies, data modeling, or cloud database selection. Cloud vendor guide absorbed from moai-platform-database-cloud.

Frontend development specialist covering React 19, Next.js 16, Vue 3.5, and modern UI/UX patterns with component architecture. Use when building web UIs, implementing components, optimizing frontend performance, or integrating state management.

Markdown-to-single-file-HTML report renderer. Six modes (status, incident, plan, explainer, financial, pr) selected automatically by report type. Zero external JS/CSS framework dependencies — inline SVG charts, single font-CDN exception for Korean readability. Self-contained output for email attachment, print, and offline viewing.

21 hooks across 20 events

gopls

pyright-langserver

rust-analyzer

sourcekit-lsp

typescript-language-server

Personal technical learning tutor grounded in official documentation via Context7 MCP. Explains concepts using analogies, generates markdown study notes with Mermaid diagrams in .moai/learning/, and optionally syncs lectures to Notion for mobile learning. Audits your understanding instead of just feeding answers.

Beginner-friendly pair programming companion — a simplified, approachable version of MoAI for first-time developers and coding newcomers. Keeps MoAI's four-step rhythm (Understand → Plan → Do → Check) but explains every technical term in plain words, guides step-by-step at a learner's pace, and adds a warm mentor tone. Writes real code AND stays friendly. Delegates to specialists when it genuinely helps — and always explains why.

Agentic coding orchestrator that merges strategic delegation with pair programming collaboration. Clarifies intent via Socratic inquiry, delegates to specialists, gates every change through checkpoint verification, and prevents dark-flow over-engineering. Built for long-horizon multi-hour coding sessions.