惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

H
Help Net Security
T
ThreatConnect
SecWiki News
SecWiki News
F
Future of Privacy Forum
AWS News Blog
AWS News Blog
C
Cisco Blogs
A
Arctic Wolf
Vercel News
Vercel News
The GitHub Blog
The GitHub Blog
Scott Helme
Scott Helme
V
V2EX
博客园 - 叶小钗
阮一峰的网络日志
阮一峰的网络日志
K
Kaspersky official blog
G
Google Developers Blog
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
P
Privacy International News Feed
C
Cyber Attacks, Cyber Crime and Cyber Security
N
News | PayPal Newsroom
Schneier on Security
Schneier on Security
NISL@THU
NISL@THU
Microsoft Azure Blog
Microsoft Azure Blog
量子位
The Hacker News
The Hacker News
Stack Overflow Blog
Stack Overflow Blog
Security Latest
Security Latest
M
Microsoft Research Blog - Microsoft Research
Google Online Security Blog
Google Online Security Blog
博客园_首页
C
CXSECURITY Database RSS Feed - CXSecurity.com
I
InfoQ
Google DeepMind News
Google DeepMind News
Y
Y Combinator Blog
The Cloudflare Blog
Microsoft Security Blog
Microsoft Security Blog
Martin Fowler
Martin Fowler
Cisco Talos Blog
Cisco Talos Blog
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
T
Troy Hunt's Blog
F
Fox-IT International blog
S
Security @ Cisco Blogs
博客园 - 司徒正美
cs.CV updates on arXiv.org
cs.CV updates on arXiv.org
C
Comments on: Blog
Threat Intelligence Blog | Flashpoint
Threat Intelligence Blog | Flashpoint
L
LINUX DO - 最新话题
GbyAI
GbyAI
Project Zero
Project Zero
腾讯CDC
T
Tailwind CSS Blog

DEV Community

How I Built 100 Browser-Based Image Tools With No Server (FFmpeg WASM, PDF-lib, AI Background Removal) Nginx CVE-2026-9256, AI Prompt Injection Defenses, and Claude AI Data Leak Demo Scaling RAG for 10M+ Docs, .md Agent Memory, & Claude Code for Motion Graphics Diagram as Code with draw.io DuckDB Delta, PostgreSQL 17 Migration, & SQLite Optimization Deep Dives Windows 11 Microsoft Account Login Recovery During Internet Restrictions The Linux Commands You Forgot Exist (And Why AI Workflows Make Them Relevant Again) Spec-Driven Development Without an IDE: I Generated NestJS, Go, Spring Boot, Laravel, and Rust Apps From a Single PRD File Components are states Edge SEO y Middleware: Cómo Interceptar a Googlebot y LLMs antes de llegar a tu Servidor Context window exceeded at turn 23. Here's how I track token usage without a tokenizer. My Hermes agent spent $3 before I noticed. Now it can't. My Hermes agent's stop condition was a 40-line if/elif chain. I replaced it with 3 lines. My agent kept hitting context limits. This one function fixed it. Create and configure Azure Firewall Your Hermes agent's audit log is leaking customer emails. Here's a 100-line lib that fixes that. My agent kept forgetting what it was doing. A scratchpad fixed it. I replaced 200 lines of ad-hoc state management in my Hermes agent with one object. Per-Key Rate Limiting for Agent Tool Calls: Stop One User From Breaking Everything Composable Output Guardrails: Filter Agent Responses Before They Reach Users Sanitize Your LLM Message Lists Before Every API Call Thread a Run ID Through Every Agent Call So You Can Debug Anything Normalize Provider Error JSON So Your Agent Can Actually Handle Failures Priority Queue for Agent Sub-Tasks: Stop Processing Low-Priority Work First Static Lint Rules for Your LLM Prompts (Before They Hit Production) tool-call-budgets: Stop Runaway Agent Loops Before They Hit Your Invoice Step Through Your Agent's Failures Like a Debugger The Simplest Stop Condition: A Hard Cap on Agent Loop Iterations Score Your Agent's Responses With a 0.0-1.0 Rubric (No LLM Judge Required) Fix Bad Structured Output by Feeding the Error Back to the Model Building an effective Storyblok Tool Plugin with SvelteKit How to Get Your Renault / Dacia Radio Code for Free RAG 시스템 실전 구축 (v39) Retraction — scrml’s Living Compiler I built a fitness app where the AI roasts you for eating pizza (and hypes you when you PR) The Top SaaS Founder Communities on Discord (Beyond the AI Hype) I Built a Production-Grade Async Job Queue from Scratch — Here's Everything That Actually Happened How to watch SMS from multiple Android phones in one iOS app We Didn’t Want Another AI Wrapper — So We Explored a High-Speed Hermes Orchestrator for Engineering Crews Multi-tenant além do TenantId: problemas reais e aprendizados em sistemas .NET After failing 23 times, I am sharing How I Actually Prepare for a Tech Interview Every Single Time Now. I built an app that works like a nutritionist for your brain. Here's what happened in 7 days. GoBadge Dynamic: From Module Stats to Universal Badges LangGraph 워크플로우 템플릿 (v39) The git Commands You Forgot Exist (And Why AI Workflows Make Them Relevant Again) Six Levels of MCP Servers One container to replace Grafana + Loki + Tempo + Prometheus The Request/Response Cycle, HTTP, Auth, JWT, OAuth & Sessions — Explained Properly Python Week 3: We Stopped Repeating Ourselves (Loops!) Creating a Custom Grid Editor tool in Unreal Engine 我做了个付费 Telegram bot。Telegram Stars 实际给开发者多少钱,我算了一笔账。 I Got 96% Recall on LLM Hallucination Detection With No ML Model – Just 50 Lines of Python A practitioner's guide to getting more value out of AI coding: agent quality & token optimization How to Handle Telegram Albums in Telegraf I Built a Multilingual Spam Detection Dataset with 149K+ Messages Across 23 Languages How to Handle Telegram Albums in grammY RAG 시스템 실전 구축 (v38) Beyond Pip Install: Why Your AI Agent Needs a "Hermetic" Life-Support System to Survive Resume Building using HTML & CSS SpecFlow: Multi-Agent SDD in Cursor (4 phases, /approve, single code writer) Running ASR for smart homes in the NPU of Intel processors "Building a CI/CD Pipeline From Scratch: A Practical Guide for Developers (with GitHub Actions)" SpecFlow: SDD multi-agente en Cursor (4 fases, /approve, un solo escritor de código) How to Extract Your Full Team Hierarchy from HubSpot (the API doesn't expose it) Adobe Commerce Cloud now costs $40k/year. We migrated from Adobe Commerce to Magento Open Source — here's the honest breakdown .klickd v4.0.0 — Portable AI memory with constraints, strict schemas, and test vectors We Trust Third Party Code, It’s Time to Trust AI Generated Code LangGraph 워크플로우 템플릿 (v38) Sustainable AI Starts with Efficient AI Find Remove duplicated files in Google Drive How to Detect GPU Waste in a Kubernetes Cluster The Privacy Bug in My First Chrome Extension (And How to Avoid It) Serverless Mental Models: What They Don't Tell You Before You Build Preventing GPT hallucination in automated content pipelines: how I structure Make.com flows with data injection Hmm, where were we? AI Visibility Tools, Math Proofs, and Stripped Guardrails Shape Developer Landscape How AI and Electronics Are Changing Healthcare Devices: The Future of Smart Healthcare Author: Shivam Wakade | Founder, PrivSR Making Claude Sound Like Optimus Prime Understanding Reinforcement Learning with Human Feedback Part 5: Training the Reward Model with Loss Functions Learning Progress Pt.20 How Secure LoRa Communication Devices Work: Building the Future of Private and Long-Range Connectivity Author: Shivam Wakade | Founder, PrivSR How I Rebuilt an RPG Map Editor with Rust, React, and WASM Building a System That Automates YouTube Post-Production Building a 100% Serverless Digital Asset Packager in the Browser Game Recommended AI What is Human-In-The-Loop (HITL)? Deep Dive: React Server Components in TanStack Start Migrating off Google Analytics: Umami vs Plausible vs Fathom Building a Portfolio That Actually Demonstrates Software Engineering Async/Await in JavaScript: From Callbacks to Clean Code (2026) Benchmarking LLM Structured Outputs Angular 21 Multiselect Dropdown: A Migration-Friendly Component with Live Functional Tests ShareBox v5 — GPU transcoding, Netflix-style grid, and why I don't need Plex anymore TOML Schema is live Handling Duplicate Shopify Webhook Events (And Why You Must) Original Kubernetes Dashboard — retired upstream, upgraded to Angular 21. لماذا أسست ترينافو للتجار العرب الذين تتجاهلهم المنصات الغربية Construyendo un recomendador de películas en Python: de los datos al modelo When APIs Lie: A Lesson in Defensive Debugging Pope Leo XIV's AI Encyclical: What Builders Must Know (2026)
LangGraph 워크플로우 템플릿 (v40)
matias yoon · 2026-05-26 · via DEV Community

LangGraph 워크플로우 템플릿 (v40)

1. LangGraph 아키텍처 개요

LangGraph는 상태 기반의 워크플로우 엔진으로, 노드(Node), 엣지(Edge), 상태(State), 체크포인트(Checkpointing)로 구성됩니다.

from langgraph.graph import StateGraph, END
from langgraph.checkpoint.memory import MemorySaver
from typing import TypedDict, Annotated
import operator

# 상태 정의
class AgentState(TypedDict):
    messages: Annotated[list, operator.add]
    next: str

# 워크플로우 정의
workflow = StateGraph(AgentState)
workflow.add_node("retrieve", retrieve_node)
workflow.add_node("generate", generate_node)
workflow.add_edge("retrieve", "generate")
workflow.add_edge("generate", END)
workflow.set_entry_point("retrieve")

Enter fullscreen mode Exit fullscreen mode

2. 템플릿 1: 간단한 RAG 에이전트 (검색 → 생성 → 검증)

실제 데이터베이스와 LLM을 통합하는 데 필요한 최소한의 RAG 구조입니다:

import sqlite3
from langchain_core.messages import HumanMessage
from langchain_openai import ChatOpenAI

class SimpleRAGAgent:
    def __init__(self, db_path: str):
        self.db_path = db_path
        self.llm = ChatOpenAI(model="gpt-4o-mini")

    def retrieve(self, state: AgentState) -> AgentState:
        messages = state["messages"]
        query = messages[-1].content

        # 데이터베이스에서 관련 문서 검색
        conn = sqlite3.connect(self.db_path)
        cursor = conn.cursor()
        cursor.execute("""
            SELECT content FROM documents 
            WHERE content LIKE ? 
            ORDER BY similarity(?, content) DESC 
            LIMIT 5
        """, (f"%{query}%", query))

        docs = cursor.fetchall()
        conn.close()

        return {"documents": [doc[0] for doc in docs]}

    def generate(self, state: AgentState) -> AgentState:
        # 검색된 문서와 질문으로 생성
        docs = state["documents"]
        messages = state["messages"]

        prompt = f"""
        문서 내용: {docs}
        질문: {messages[-1].content}
        답변:
        """

        response = self.llm.invoke([HumanMessage(content=prompt)])
        return {"messages": [response]}

    def validate(self, state: AgentState) -> AgentState:
        # 생성된 답변 검증
        response = state["messages"][-1].content
        if len(response) < 10:
            return {"messages": [HumanMessage(content="답변이 너무 짧습니다. 다시 시도해주세요.")]}
        return {"next": "continue"}

# 워크플로우 구축
workflow = StateGraph(AgentState)
workflow.add_node("retrieve", SimpleRAGAgent.retrieve)
workflow.add_node("generate", SimpleRAGAgent.generate)
workflow.add_node("validate", SimpleRAGAgent.validate)
workflow.add_edge("retrieve", "generate")
workflow.add_edge("generate", "validate")
workflow.add_edge("validate", END)
workflow.set_entry_point("retrieve")

Enter fullscreen mode Exit fullscreen mode

3. 템플릿 2: 멀티-도구 에이전트 (계획 → 실행 → 관찰 → 결정)

다수의 도구를 활용하여 복잡한 작업을 수행하는 에이전트:

from langchain.tools import Tool
from langchain_core.tools import tool

# 도구 정의
@tool
def search_web(query: str) -> str:
    """웹 검색 도구"""
    return f"검색 결과: {query}에 대한 정보를 찾았습니다"

@tool
def calculate_math(expression: str) -> str:
    """수학 계산 도구"""
    try:
        result = eval(expression)
        return f"결과: {result}"
    except:
        return "계산 오류"

@tool
def save_data(key: str, value: str) -> str:
    """데이터 저장 도구"""
    return f"데이터 저장: {key} = {value}"

class MultiToolAgent:
    def __init__(self):
        self.tools = [search_web, calculate_math, save_data]
        self.llm = ChatOpenAI(model="gpt-4o-mini")

    def plan(self, state: AgentState) -> AgentState:
        # 현재 상태에서 필요한 작업 계획
        messages = state["messages"]
        task = messages[-1].content

        prompt = f"""
        작업: {task}
        다음 작업을 계획하세요:
        1. 어떤 도구가 필요한가?
        2. 실행 순서는?
        3. 예상 결과는?
        """

        plan = self.llm.invoke([HumanMessage(content=prompt)])
        return {"plan": plan.content}

    def execute(self, state: AgentState) -> AgentState:
        # 계획된 작업 실행
        plan = state["plan"]
        # 간단한 도구 실행 구현
        return {"execution_results": "실행 완료"}

    def observe(self, state: AgentState) -> AgentState:
        # 실행 결과 관찰
        results = state["execution_results"]
        return {"observation": f"관찰 결과: {results}"}

    def decide(self, state: AgentState) -> AgentState:
        # 다음 단계 결정
        observation = state["observation"]
        if "실패" in observation:
            return {"next": "retry"}
        return {"next": "continue"}

# 워크플로우 구축
workflow = StateGraph(AgentState)
workflow.add_node("plan", MultiToolAgent.plan)
workflow.add_node("execute", MultiToolAgent.execute)
workflow.add_node("observe", MultiToolAgent.observe)
workflow.add_node("decide", MultiToolAgent.decide)
workflow.add_edge("plan", "execute")
workflow.add_edge("execute", "observe")
workflow.add_edge("observe", "decide")
workflow.add_edge("decide", END)
workflow.set_entry_point("plan")

Enter fullscreen mode Exit fullscreen mode

4. 템플릿 3: 인간-중개 워크플로우 (일시정지 → 검토 → 계속)

사람의 검토를 포함한 작업 흐름:

import time
from typing import Literal

class HumanInLoopAgent:
    def __init__(self, model="gpt-4o-mini"):
        self.llm = ChatOpenAI(model=model)
        self.checkpoint = None

    def process(self, state: AgentState) -> AgentState:
        # AI 처리
        messages = state["messages"]
        prompt = f"처리 요청: {messages[-1].content}"
        response = self.llm.invoke([HumanMessage(content=prompt)])
        return {"ai_response": response.content}

    def pause_for_review(self, state: AgentState) -> AgentState:
        # 인간 검토를 위한 일시정지
        ai_response = state["ai_response"]
        return {
            "status": "waiting_for_review",
            "review_request": f"검토 요청:\n{ai_response}"
        }

    def continue_workflow(self, state: AgentState) -> AgentState:
        # 검토 후 계속 진행
        if state["review_approved"]:
            return {"messages": [HumanMessage(content="검토 승인")]}
        return {"messages": [HumanMessage(content="검토 거부")]}

    def retry_with_feedback(self, state: AgentState) -> AgentState:
        # 피드백 기반 재시도
        feedback = state["feedback"]
        return {"messages": [HumanMessage(content=f"피드백 반영: {feedback}")]}

    def wait_for_human_input(self) -> AgentState:
        # 인간 입력 대기
        print("사람의 입력을 기다리는 중...")
        human_input = input("사람 입력: ")
        return {"messages": [HumanMessage(content=human_input)]}

# 인간 검토 워크플로우
workflow = StateGraph(AgentState)
workflow.add_node("process", HumanInLoopAgent.process)
workflow.add_node("pause", HumanInLoopAgent.pause_for_review)
workflow.add_node("continue", HumanInLoopAgent.continue_workflow)
workflow.add_edge("process", "pause")
workflow.add_edge("pause", "continue")
workflow.add_edge("continue", END)
workflow.set_entry_point("process")

Enter fullscreen mode Exit fullscreen mode

5. 템플릿 5: 병렬 실행 에이전트 (분기 → 처리 → 집계)

여러 작업을 동시에 처리 후 결과 집계:


python
from concurrent.futures import ThreadPoolExecutor
import asyncio

class ParallelAgent:
    def __init__(self, max_workers=4):
        self.max_workers = max_workers
        self.llm = ChatOpenAI(model="gpt-4o-mini")

    def fan_out(self, state: AgentState) -> AgentState:
        # 작업 분기
        messages = state["messages"]
        task = messages[-1].content

        # 병렬 처리를 위한 작업 목록 생성
        tasks = [
            f"분석: {task} - 1",
            f"분석: {task} - 2", 
            f"분석: {task} - 3"
        ]
        return {"parallel_tasks":

---

📥 **Get the full guide on Gumroad**: https://gumroad.com/l/auto ($5)

Enter fullscreen mode Exit fullscreen mode