Stats
Actions
Tags
Help us improve
Share bugs, ideas, or general feedback.
Use when building or extending LangGraph agent systems with layered architecture, creating StateGraph projects, scaffolding 7-layer structures (core, memory, prompts, tools, nodes, graphs, interfaces, config), adding tools to registry, creating subgraphs, generating nodes by role type (reasoning, execution, routing), managing prompt versions, or running dependency lint.
npx claudepluginhub jhleee/layered-agent-architecture --plugin layered-agent-architectHow this skill is triggered — by the user, by Claude, or both
Slash command
/layered-agent-architect:layered-agent-architectThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
이 스킬은 LangGraph 기반 에이전트 시스템의 7-레이어 아키텍처를 구축·확장할 때 사용한다.
assets/architecture-rules.mdexamples/01-scaffold-phase1.mdexamples/02-add-tool.mdexamples/03-add-subgraph-and-node.mdreferences/01-scaffold.mdreferences/02-tool-creator.mdreferences/03-subgraph-creator.mdreferences/04-node-creator.mdreferences/05-prompt-versioning.mdreferences/06-docs-generator.mdreferences/07-dependency-lint.mdscaffolding/config/__init__.pyscaffolding/config/agents.yamlscaffolding/config/settings.pyscaffolding/core/__init__.pyscaffolding/core/models.pyscaffolding/core/schemas.pyscaffolding/core/state.pyscaffolding/graphs/__init__.pyscaffolding/graphs/builder.pyBuilds production-grade stateful multi-actor AI agents with LangGraph, covering graph construction, state management, persistence, cycles, branches, human-in-the-loop, and ReAct patterns.
Builds production-grade AI agents with LangGraph: graph construction, state management, persistence, human-in-the-loop, and the ReAct agent pattern.
Guides architectural decisions for LangGraph applications. Use when choosing between LangGraph vs alternatives, designing state schemas with reducers, structuring graphs with subgraphs, or selecting persistence and streaming approaches.
Share bugs, ideas, or general feedback.
이 스킬은 LangGraph 기반 에이전트 시스템의 7-레이어 아키텍처를 구축·확장할 때 사용한다.
assets/architecture-rules.md를 먼저 읽는다scaffolding/ 디렉토리를 복사한다 — 코드 생성 대신 복사 후 수정| 사용자 의도 | 참조 파일 | 설명 |
|---|---|---|
| 프로젝트 생성, 스케폴딩, 초기화 | references/01-scaffold.md | scaffolding/ 복사 + 커스터마이즈 |
| 도구 추가, tool 생성 | references/02-tool-creator.md | @tool + 레지스트리 등록 |
| 서브그래프 추가, 멀티에이전트 | references/03-subgraph-creator.md | 부분 State + 팩토리 패턴 |
| 노드 추가, reasoning/execution/routing | references/04-node-creator.md | 역할별 코드 템플릿 |
| 프롬프트 버전, A/B 테스트 | references/05-prompt-versioning.md | 버전 관리 + builder 분기 |
| 문서 생성, README, 다이어그램 | references/06-docs-generator.md | Mermaid + README 템플릿 |
| 의존성 검증, lint, 규칙 검사 | references/07-dependency-lint.md | AST 기반 import 분석 |
src/
├── core/ # State, Schema, Model Factory
├── memory/ # Checkpointer, Store
├── prompts/ # Templates, Builder
├── tools/ # Registry, 개별 도구
├── nodes/ # reasoning, execution, routing
├── graphs/ # main, builder, 서브그래프
├── interfaces/ # API (FastAPI), Streaming (SSE)
├── config/ # settings.py, agents.yaml
└── main.py # 진입점
main.py
└→ interfaces/
└→ graphs/
├→ nodes/
│ ├→ prompts/
│ ├→ tools/
│ └→ core/ ← 최하위
├→ memory/
└→ config/ ← 수평 참조 (어디서든 import 가능)