G\u00fcn\u00fcm\u00fcz\u00fcn h\u0131zla geli\u015fen teknoloji d\u00fcnyas\u0131nda, b\u00fcy\u00fck dil modellerinin (LLM) g\u00fcc\u00fcn\u00fc kullanarak karma\u015f\u0131k ara\u015ft\u0131rma g\u00f6revlerini otomatikle\u015ftirmek, bilgiye eri\u015fimi ve analizi k\u00f6kten de\u011fi\u015ftirmektedir. Ancak, elde edilen verilerin sadece metin tabanl\u0131 sunumu \u00e7o\u011fu zaman yetersiz kal\u0131r. Dinamik ve etkile\u015fimli grafiklerle zenginle\u015ftirilmi\u015f \u00e7\u0131kt\u0131lar, verilerin daha h\u0131zl\u0131 kavranmas\u0131n\u0131, trendlerin kolayca tespit edilmesini ve karar alma s\u00fcre\u00e7lerinin iyile\u015ftirilmesini sa\u011flar. Bu makalede, LangGraph \u00e7at\u0131s\u0131 alt\u0131nda g\u00fc\u00e7l\u00fc bir ara\u015ft\u0131rma ajan\u0131 olu\u015fturmay\u0131 ve bu ajan\u0131n analiz sonu\u00e7lar\u0131n\u0131 CopilotKit ve Tako gibi MCP (Microsoft Copilot Platform) uygulamalar\u0131 arac\u0131l\u0131\u011f\u0131yla dinamik grafiklerle nas\u0131l g\u00f6rselle\u015ftirece\u011fini ad\u0131m ad\u0131m inceleyece\u011fiz. Amac\u0131m\u0131z, hem verimli bir ara\u015ft\u0131rma s\u00fcreci sunan hem de g\u00f6rsel olarak zengin, etkile\u015fimli \u00e7\u0131kt\u0131lar \u00fcreten entegre bir \u00e7\u00f6z\u00fcm geli\u015ftirmektir.<\/p>\n
Yapay zeka tabanl\u0131 sistemlerin karma\u015f\u0131k g\u00f6revleri yerine getirme yetene\u011fi, \u00f6zellikle b\u00fcy\u00fck dil modellerinin (LLM) y\u00fckseli\u015fiyle birlikte \u00f6nemli \u00f6l\u00e7\u00fcde artt\u0131. Bu sistemlerin otonom bir \u015fekilde ara\u015ft\u0131rma yapabilmesi, veri toplayabilmesi ve analiz edebilmesi, bir\u00e7ok sekt\u00f6r i\u00e7in devrim niteli\u011findedir. \u0130\u015fte bu noktada LangGraph ve ara\u015ft\u0131rma ajanlar\u0131 devreye giriyor.<\/p>\n
LangGraph, LangChain ekosisteminin bir par\u00e7as\u0131 olarak, \u00e7ok ad\u0131ml\u0131 LLM uygulamalar\u0131 olu\u015fturmak i\u00e7in tasarlanm\u0131\u015f bir k\u00fct\u00fcphanedir. Geleneksel LangChain zincirlerinin aksine, LangGraph durum bilgisi olan (stateful) ajanlar ve d\u00f6ng\u00fcler olu\u015fturman\u0131za olanak tan\u0131r. Bu, ajanlar\u0131n \u00f6nceki ad\u0131mlardan gelen bilgileri hat\u0131rlamas\u0131n\u0131, karar vermesini ve belirli ko\u015fullar alt\u0131nda tekrar eden g\u00f6revleri yerine getirmesini sa\u011flar. Bir grafik yap\u0131s\u0131 kullanarak, farkl\u0131 d\u00fc\u011f\u00fcmler (LLM \u00e7a\u011fr\u0131lar\u0131, ara\u00e7 kullan\u0131mlar\u0131, insan m\u00fcdahaleleri) aras\u0131nda ge\u00e7i\u015fleri tan\u0131mlayabilir ve b\u00f6ylece karma\u015f\u0131k i\u015f ak\u0131\u015flar\u0131n\u0131 modelleyebilirsiniz.<\/p>\n
Ara\u015ft\u0131rma ajanlar\u0131, belirli bir konuda bilgi toplamak, analiz etmek ve \u00f6zetlemek i\u00e7in tasarlanm\u0131\u015f otonom sistemlerdir. Geleneksel olarak insanlar taraf\u0131ndan yap\u0131lan bu g\u00f6revler, ajanlar sayesinde \u00e7ok daha h\u0131zl\u0131, kapsaml\u0131 ve tutarl\u0131 bir \u015fekilde ger\u00e7ekle\u015ftirilebilir. Bir ara\u015ft\u0131rma ajan\u0131, internette arama yapabilir, belgeleri okuyabilir, veritabanlar\u0131n\u0131 sorgulayabilir ve elde etti\u011fi bilgileri sentezleyerek belirli bir soruya yan\u0131t verebilir veya bir rapor olu\u015fturabilir. Bu ajanlar, finans, sa\u011fl\u0131k, hukuk ve teknoloji gibi bir\u00e7ok alanda de\u011ferli bilgiler sa\u011flayabilir.<\/p>\n
LangGraph, ara\u015ft\u0131rma ajanlar\u0131 geli\u015ftirmek i\u00e7in ideal bir platform sunar \u00e7\u00fcnk\u00fc:<\/p>\n
Veri, g\u00fcn\u00fcm\u00fcz\u00fcn en de\u011ferli varl\u0131klar\u0131ndan biridir. Ancak ham veriler, genellikle karma\u015f\u0131k ve anla\u015f\u0131lmas\u0131 zordur. \u0130\u015fte bu noktada veri g\u00f6rselle\u015ftirme devreye girer.<\/p>\n
Veri g\u00f6rselle\u015ftirme, karma\u015f\u0131k veri k\u00fcmelerini grafikler, tablolar ve di\u011fer g\u00f6rsel \u00f6\u011feler arac\u0131l\u0131\u011f\u0131yla daha anla\u015f\u0131l\u0131r hale getirme sanat\u0131d\u0131r. Bu, insanlar\u0131n b\u00fcy\u00fck veri hacimlerindeki kal\u0131plar\u0131, trendleri ve ayk\u0131r\u0131 de\u011ferleri daha h\u0131zl\u0131 ve etkili bir \u015fekilde tan\u0131mas\u0131na yard\u0131mc\u0131 olur. \u0130yi bir g\u00f6rselle\u015ftirme, hikaye anlat\u0131r, i\u00e7g\u00f6r\u00fcler sunar ve karar verme s\u00fcre\u00e7lerini destekler.<\/p>\n
Bir ara\u015ft\u0131rma ajan\u0131, b\u00fcy\u00fck miktarda veri toplay\u0131p analiz etti\u011finde, bu verilerin sadece metin olarak sunulmas\u0131 yetersiz kalabilir. Dinamik grafikler, ajan\u0131n elde etti\u011fi i\u00e7g\u00f6r\u00fcleri \u00e7ok daha etkili bir \u015fekilde aktarabilir:<\/p>\n
Microsoft Copilot Platform (MCP) uygulamalar\u0131, yapay zeka destekli deneyimler olu\u015fturmak i\u00e7in g\u00fc\u00e7l\u00fc ara\u00e7lar sunar. Bu makalede, dinamik grafik entegrasyonu i\u00e7in CopilotKit ve Tako’nun birle\u015fimine odaklanaca\u011f\u0131z.<\/p>\n
CopilotKit, geli\u015ftiricilerin uygulamalar\u0131na “copilot” benzeri yetenekler eklemesini sa\u011flayan bir a\u00e7\u0131k kaynakl\u0131 \u00e7er\u00e7evedir. Kullan\u0131c\u0131lar\u0131n do\u011fal dil kullanarak uygulamalarla etkile\u015fim kurmas\u0131na, g\u00f6revleri otomatikle\u015ftirmesine ve i\u00e7erik olu\u015fturmas\u0131na olanak tan\u0131r. \u00d6zellikle React tabanl\u0131 \u00f6n u\u00e7 uygulamalar\u0131 i\u00e7in tasarlanm\u0131\u015ft\u0131r ve LLM’leri, ara\u00e7lar\u0131 ve kullan\u0131c\u0131 aray\u00fcz\u00fcn\u00fc sorunsuz bir \u015fekilde birle\u015ftirir. Bir LangGraph ajan\u0131 taraf\u0131ndan \u00fcretilen dinamik grafikleri kullan\u0131c\u0131 aray\u00fcz\u00fcnde g\u00f6stermek i\u00e7in CopilotKit’in sundu\u011fu esneklik paha bi\u00e7ilmezdir.<\/p>\n
Tako, Microsoft taraf\u0131ndan geli\u015ftirilen ve \u00f6zellikle CopilotKit ile entegre \u00e7al\u0131\u015fmak \u00fczere tasarlanm\u0131\u015f, dinamik ve etkile\u015fimli grafikler olu\u015fturmaya y\u00f6nelik bir k\u00fct\u00fcphanedir. Veri g\u00f6rselle\u015ftirmeyi basitle\u015ftirir ve geli\u015ftiricilerin karma\u015f\u0131k grafikler olu\u015fturmas\u0131n\u0131 sa\u011flar. Tako’nun g\u00fcc\u00fc, verileri kolayca g\u00f6rselle\u015ftirilebilir formatlara d\u00f6n\u00fc\u015ft\u00fcrme ve bu grafikleri web uygulamalar\u0131na g\u00f6mme yetene\u011finden gelir. Bir ara\u015ft\u0131rma ajan\u0131 taraf\u0131ndan analiz edilen verileri, anla\u015f\u0131l\u0131r ve etkile\u015fimli grafiklere d\u00f6n\u00fc\u015ft\u00fcrmek i\u00e7in ideal bir ara\u00e7t\u0131r.<\/p>\n
CopilotKit ve Tako’nun birle\u015fimi, LangGraph ara\u015ft\u0131rma ajan\u0131n\u0131z i\u00e7in g\u00fc\u00e7l\u00fc bir g\u00f6rselle\u015ftirme katman\u0131 olu\u015fturur:<\/p>\n
Etkili bir LangGraph ara\u015ft\u0131rma ajan\u0131 tasarlamak, farkl\u0131 bile\u015fenlerin nas\u0131l etkile\u015fim kuraca\u011f\u0131n\u0131 ve veri ak\u0131\u015f\u0131n\u0131n nas\u0131l y\u00f6netilece\u011fini dikkatlice planlamay\u0131 gerektirir.<\/p>\n
Bir ara\u015ft\u0131rma ajan\u0131 i\u00e7in tipik bir ak\u0131\u015f a\u015fa\u011f\u0131daki gibi olabilir:<\/p>\n
Dinamik grafikler, ajan\u0131n ak\u0131\u015f\u0131n\u0131n farkl\u0131 noktalar\u0131na entegre edilebilir:<\/p>\n
\u015eimdi, LangGraph kullanarak temel bir ara\u015ft\u0131rma ajan\u0131 olu\u015fturma s\u00fcrecini inceleyelim. Bu \u00f6rnekte, basit bir web arama ve \u00f6zetleme ajan\u0131 geli\u015ftirece\u011fiz.<\/p>\n
\u00d6ncelikle gerekli k\u00fct\u00fcphaneleri y\u00fcklememiz gerekiyor:<\/p>\n
pip install langchain langgraph langchain-openai langchain-community tavily-python copilotkit-py takos<\/pre>\n<\/code><\/p>\n
API anahtarlar\u0131n\u0131z\u0131 ortam de\u011fi\u015fkenleri olarak ayarlad\u0131\u011f\u0131n\u0131zdan emin olun (\u00f6rne\u011fin, OPENAI_API_KEY<\/code>, TAVILY_API_KEY<\/code>).<\/p>\nLangGraph Temel Yap\u0131s\u0131 ve D\u00fc\u011f\u00fcm Tan\u0131mlar\u0131<\/h4>\n
Bir LangGraph ajan\u0131, bir dizi d\u00fc\u011f\u00fcm ve bu d\u00fc\u011f\u00fcmler aras\u0131ndaki ge\u00e7i\u015flerden olu\u015fur. Her d\u00fc\u011f\u00fcm bir i\u015flevi temsil eder.<\/p>\n
from langgraph.graph import StateGraph, END\nfrom langchain_openai import ChatOpenAI\nfrom langchain_community.tools import TavilySearchResults\nfrom typing import TypedDict, Annotated, List\nimport operator\n\n# 1. Ajan\u0131n Durumunu Tan\u0131mlama\nclass AgentState(TypedDict):\n query: str\n search_results: List[str]\n analysis_report: str\n chart_data: dict # Tako i\u00e7in grafik verileri\n\n# 2. Ara\u00e7lar\u0131 Tan\u0131mlama\nsearch_tool = TavilySearchResults(max_results=5) # Web arama arac\u0131\n\n# 3. LLM'i Tan\u0131mlama\nllm = ChatOpenAI(model=\"gpt-4o\", temperature=0)\n\n# 4. D\u00fc\u011f\u00fcm Fonksiyonlar\u0131n\u0131 Tan\u0131mlama\n\ndef execute_search(state: AgentState):\n print(\"---ARAMA YAPILIYOR---\")\n query = state[\"query\"]\n results = search_tool.invoke({\"query\": query})\n return {\"search_results\": [str(r) for r in results]}\n\ndef analyze_results(state: AgentState):\n print(\"---SONU\u00c7LAR ANAL\u0130Z ED\u0130L\u0130YOR---\")\n search_results = state[\"search_results\"]\n query = state[\"query\"]\n\n prompt = f\"\"\"\n A\u015fa\u011f\u0131daki arama sonu\u00e7lar\u0131n\u0131 '{query}' konusu \u00f6zelinde analiz et ve \u00f6zetle.\n \u00d6zellikle \u00f6nemli noktalar\u0131, trendleri ve olas\u0131 \u00e7\u0131kar\u0131mlar\u0131 belirt.\n Analizini yaparken, daha sonra grafik olu\u015fturulabilecek say\u0131sal veriler veya kar\u015f\u0131la\u015ft\u0131rmalar varsa,\n bunlar\u0131 belirt ve yap\u0131land\u0131r\u0131lm\u0131\u015f bir \u015fekilde sunmaya \u00e7al\u0131\u015f.\n\n Arama Sonu\u00e7lar\u0131:\n {'\\n'.join(search_results)}\n \"\"\"\n analysis = llm.invoke(prompt).content\n return {\"analysis_report\": analysis}\n\ndef generate_chart_data(state: AgentState):\n print(\"---GRAF\u0130K VER\u0130S\u0130 HAZIRLANIYOR---\")\n analysis_report = state[\"analysis_report\"]\n query = state[\"query\"]\n\n # LLM'den grafik i\u00e7in uygun veri format\u0131n\u0131 isteme\n # \u00d6rnek: \"2023 y\u0131l\u0131ndaki gelir da\u011f\u0131l\u0131m\u0131\" gibi bir analizden\n # y\u0131llara g\u00f6re gelir verilerini \u00e7\u0131karmas\u0131n\u0131 isteyebiliriz.\n chart_prompt = f\"\"\"\n A\u015fa\u011f\u0131daki analiz raporundan, '{query}' konusuyla ilgili g\u00f6rselle\u015ftirilebilecek anahtar say\u0131sal verileri\n bir JSON format\u0131nda \u00e7\u0131kar. \u00d6rne\u011fin, e\u011fer rapor farkl\u0131 kategoriler veya zaman dilimleri aras\u0131nda\n kar\u015f\u0131la\u015ft\u0131rmalar i\u00e7eriyorsa, bunlar\u0131 bir liste veya s\u00f6zl\u00fck olarak yap\u0131land\u0131r.\n \u00d6rnek JSON format\u0131:\n {{\n \"chart_type\": \"bar\",\n \"title\": \"{query} Konusunda Anahtar Veriler\",\n \"x_axis\": \"Kategori\",\n \"y_axis\": \"De\u011fer\",\n \"data\": [\n {{\"Kategori\": \"A\", \"De\u011fer\": 100}},\n {{\"Kategori\": \"B\", \"De\u011fer\": 150}},\n {{\"Kategori\": \"C\", \"De\u011fer\": 75}}\n ]\n }}\n E\u011fer uygun say\u0131sal veri yoksa, bo\u015f bir JSON objesi d\u00f6nd\u00fcr: {{}}\n\n Analiz Raporu:\n {analysis_report}\n \"\"\"\n chart_json_str = llm.invoke(chart_prompt).content\n try:\n chart_data = json.loads(chart_json_str)\n except json.JSONDecodeError:\n chart_data = {} # JSON \u00e7\u00f6z\u00fcmlenemezse bo\u015f d\u00f6nd\u00fcr\n\n return {\"chart_data\": chart_data}\n\n# 5. LangGraph Grafi\u011fini Olu\u015fturma\nworkflow = StateGraph(AgentState)\n\nworkflow.add_node(\"search\", execute_search)\nworkflow.add_node(\"analyze\", analyze_results)\nworkflow.add_node(\"generate_chart_data\", generate_chart_data)\n\nworkflow.set_entry_point(\"search\")\n\nworkflow.add_edge(\"search\", \"analyze\")\nworkflow.add_edge(\"analyze\", \"generate_chart_data\")\nworkflow.add_edge(\"generate_chart_data\", END)\n\napp = workflow.compile()\n\n# Ajan\u0131 \u00e7al\u0131\u015ft\u0131rma\n# for s in app.stream({\"query\": \"2023 T\u00fcrkiye enflasyon oranlar\u0131 ve etkileri\"}):\n# print(s)\n<\/pre>\n<\/code><\/p>\n
Veri \u0130\u015fleme ve Grafik Olu\u015fturma Mant\u0131\u011f\u0131<\/h4>\n
Yukar\u0131daki \u00f6rnekte, generate_chart_data<\/code> d\u00fc\u011f\u00fcm\u00fc, LLM'den analiz raporundaki say\u0131sal verileri JSON format\u0131nda \u00e7\u0131karmas\u0131n\u0131 ister. Bu JSON, daha sonra Tako taraf\u0131ndan grafik olu\u015fturmak i\u00e7in kullan\u0131lacakt\u0131r. Bu ad\u0131m, LLM'in sadece metin de\u011fil, yap\u0131land\u0131r\u0131lm\u0131\u015f veri \u00fcretme yetene\u011fini kullan\u0131r.<\/p>\nMCP Uygulamalar\u0131 ile Dinamik Grafik Entegrasyonu<\/h3>\n
\u015eimdi, LangGraph ajan\u0131m\u0131z\u0131n \u00fcretti\u011fi grafik verilerini CopilotKit ve Tako ile nas\u0131l g\u00f6rselle\u015ftirece\u011fimizi g\u00f6relim.<\/p>\n
CopilotKit ile Kullan\u0131c\u0131 Aray\u00fcz\u00fc Olu\u015fturma<\/h4>\n
CopilotKit, React tabanl\u0131 bir \u00f6n u\u00e7 ile en iyi \u015fekilde \u00e7al\u0131\u015f\u0131r. Bir React uygulamas\u0131nda, LangGraph ajan\u0131ndan gelen yan\u0131tlar\u0131 al\u0131p CopilotKit bile\u015fenleri arac\u0131l\u0131\u011f\u0131yla g\u00f6sterebiliriz. Burada \u00f6nemli olan, LangGraph ajan\u0131n\u0131n chart_data<\/code> \u00e7\u0131kt\u0131s\u0131n\u0131 CopilotKit'e iletmek ve CopilotKit'in bu veriyi Tako'ya aktarmas\u0131n\u0131 sa\u011flamakt\u0131r.<\/p>\n\/\/ React Frontend (\u00d6rnek)\nimport React, { useState } from 'react';\nimport { CopilotKit, useCopilotContext } from '@copilotkit\/react-core';\nimport { CopilotSidebar } from '@copilotkit\/react-ui';\nimport { TakoChart } from '@copilotkit\/react-tako'; \/\/ Tako entegrasyonu\n\nfunction MyResearchApp() {\n const [researchResult, setResearchResult] = useState(null);\n const [chartData, setChartData] = useState(null);\n const copilotContext = useCopilotContext();\n\n const handleResearch = async (query) => {\n \/\/ LangGraph ajan\u0131 \u00e7a\u011fr\u0131s\u0131 (\u00f6rne\u011fin bir API endpoint \u00fczerinden)\n \/\/ Bu k\u0131s\u0131m, LangGraph ajan\u0131 backend'de \u00e7al\u0131\u015ft\u0131\u011f\u0131nda nas\u0131l \u00e7a\u011fr\u0131laca\u011f\u0131n\u0131 g\u00f6sterir.\n \/\/ Ger\u00e7ek uygulamada bir fetch iste\u011fi ile backend'deki ajana sorgu g\u00f6nderilir.\n console.log(Ara\u015ft\u0131rma ba\u015flat\u0131l\u0131yor: ${query}<\/code>);\n try {\n const response = await fetch('\/api\/research_agent', { \/\/ Kendi API endpoint'iniz\n method: 'POST',\n headers: { 'Content-Type': 'application\/json' },\n body: JSON.stringify({ query })\n });\n const data = await response.json();\n setResearchResult(data.analysis_report);\n setChartData(data.chart_data); \/\/ Ajan\u0131n d\u00f6nd\u00fcrd\u00fc\u011f\u00fc grafik verisi\n copilotContext.addChatMessages([\n { content: Ara\u015ft\u0131rma tamamland\u0131: ${query}<\/code>, role: 'assistant' },\n { content: data.analysis_report, role: 'assistant' }\n ]);\n } catch (error) {\n console.error(\"Ara\u015ft\u0131rma hatas\u0131:\", error);\n copilotContext.addChatMessages([{ content: \"Ara\u015ft\u0131rma s\u0131ras\u0131nda bir hata olu\u015ftu.\", role: 'assistant' }]);\n }\n };\n\n return (\n {\/* Kendi CopilotKit backend endpoint'iniz *\/}\n handleResearch(message.content)}\n >\n \n Ara\u015ft\u0131rma Sonu\u00e7lar\u0131<\/h3>\n {researchResult ? (\n \n {researchResult}<\/p>\n {chartData && Object.keys(chartData).length > 0 && (\n
\n G\u00f6rselle\u015ftirme<\/h4>\n
Hen\u00fcz bir ara\u015ft\u0131rma yap\u0131lmad\u0131. L\u00fctfen sa\u011fdaki asistan\u0131 kullan\u0131n.<\/p>\n )}\n <\/div>\n <\/CopilotSidebar>\n <\/CopilotKit>\n );\n}\n\nexport default MyResearchApp;\n<\/pre>\n
<\/code><\/p>\n
Tako ile Grafik Verilerini Haz\u0131rlama ve G\u00f6nderme<\/h4>\n
Yukar\u0131daki React \u00f6rne\u011finde TakoChart<\/code> bile\u015feni, LangGraph ajan\u0131ndan gelen chart_data<\/code> JSON'unu do\u011frudan al\u0131r ve g\u00f6rselle\u015ftirir. Bu, Tako'nun veri yap\u0131s\u0131n\u0131 anlad\u0131\u011f\u0131 ve otomatik olarak grafi\u011fi \u00e7izdi\u011fi anlam\u0131na gelir. Ajan\u0131n\u0131z\u0131n \u00fcretti\u011fi JSON format\u0131n\u0131n Tako'nun bekledi\u011fi formatla uyumlu olmas\u0131 \u00f6nemlidir.<\/p>\n\/\/ LangGraph ajan\u0131 taraf\u0131ndan \u00fcretilen \u00f6rnek chart_data JSON'u\n{\n \"chart_type\": \"bar\",\n \"title\": \"2023 T\u00fcrkiye Enflasyon Oranlar\u0131\",\n \"x_axis\": \"Ay\",\n \"y_axis\": \"Enflasyon Oran\u0131 (%)\",\n \"data\": [\n {\"Ay\": \"Ocak\", \"Enflasyon Oran\u0131 (%)\": 64.27},\n {\"Ay\": \"\u015eubat\", \"Enflasyon Oran\u0131 (%)\": 55.18},\n {\"Ay\": \"Mart\", \"Enflasyon Oran\u0131 (%)\": 50.51},\n {\"Ay\": \"Nisan\", \"Enflasyon Oran\u0131 (%)\": 43.68},\n {\"Ay\": \"May\u0131s\", \"Enflasyon Oran\u0131 (%)\": 39.59},\n {\"Ay\": \"Haziran\", \"Enflasyon Oran\u0131 (%)\": 38.21},\n {\"Ay\": \"Temmuz\", \"Enflasyon Oran\u0131 (%)\": 47.83},\n {\"Ay\": \"A\u011fustos\", \"Enflasyon Oran\u0131 (%)\": 58.94},\n {\"Ay\": \"Eyl\u00fcl\", \"Enflasyon Oran\u0131 (%)\": 61.53},\n {\"Ay\": \"Ekim\", \"Enflasyon Oran\u0131 (%)\": 61.36},\n {\"Ay\": \"Kas\u0131m\", \"Enflasyon Oran\u0131 (%)\": 61.98},\n {\"Ay\": \"Aral\u0131k\", \"Enflasyon Oran\u0131 (%)\": 64.77}\n ]\n}\n<\/pre>\n<\/code><\/p>\n
Bu JSON yap\u0131s\u0131, Tako'nun bir \u00e7ubuk grafik olu\u015fturmas\u0131 i\u00e7in gerekli t\u00fcm bilgileri i\u00e7erir. LLM'in bu format\u0131 do\u011fru bir \u015fekilde \u00fcretebilmesi i\u00e7in ona net talimatlar vermek kritik \u00f6neme sahiptir.<\/p>\n
Etkile\u015fimli Grafikleri Ajan Yan\u0131tlar\u0131na Dahil Etme<\/h4>\n
CopilotKit'in TakoChart<\/code> bile\u015feni sayesinde, ajan\u0131n analiziyle birlikte dinamik grafikleri do\u011frudan kullan\u0131c\u0131 aray\u00fcz\u00fcne g\u00f6mebiliriz. Kullan\u0131c\u0131lar, ajan\u0131n sundu\u011fu metin tabanl\u0131 \u00f6zetin yan\u0131 s\u0131ra, g\u00f6rselle\u015ftirmelerle de etkile\u015fim kurarak verileri daha derinlemesine ke\u015ffedebilirler. Bu, ara\u015ft\u0131rma ajan\u0131 deneyimini zenginle\u015ftirir ve sunulan bilgilerin de\u011ferini art\u0131r\u0131r.<\/p>\nOptimizasyon, Test ve Da\u011f\u0131t\u0131m \u0130pu\u00e7lar\u0131<\/h3>\n
Bir LangGraph ara\u015ft\u0131rma ajan\u0131 geli\u015ftirme s\u00fcreci sadece kod yazmaktan ibaret de\u011fildir; ayn\u0131 zamanda performans\u0131 optimize etmek, do\u011fru \u00e7al\u0131\u015ft\u0131\u011f\u0131ndan emin olmak ve \u00fcretime da\u011f\u0131t\u0131m i\u00e7in haz\u0131rlamak da \u00f6nemlidir.<\/p>\n
Performans Optimizasyonu<\/h4>\n\n- LLM Kullan\u0131m\u0131n\u0131 Optimize Edin:<\/strong> Her LLM \u00e7a\u011fr\u0131s\u0131 maliyetli ve zaman al\u0131c\u0131 olabilir. Gereksiz \u00e7a\u011fr\u0131lardan ka\u00e7\u0131n\u0131n. \u00d6nbellekleme (caching) stratejilerini de\u011ferlendirin.<\/li>\n
- Ara\u00e7 Se\u00e7imi:<\/strong> Ajan\u0131n kulland\u0131\u011f\u0131 ara\u00e7lar\u0131n (web arama motorlar\u0131 gibi) verimli oldu\u011fundan emin olun. Daha h\u0131zl\u0131 ve daha alakal\u0131 sonu\u00e7lar veren ara\u00e7lar\u0131 tercih edin.<\/li>\n
- Paralelle\u015ftirme:<\/strong> LangGraph, baz\u0131 d\u00fc\u011f\u00fcmleri paralel \u00e7al\u0131\u015ft\u0131rma yetene\u011fi sunar. M\u00fcmk\u00fcn olan yerlerde bu \u00f6zelli\u011fi kullanarak toplam \u00e7al\u0131\u015fma s\u00fcresini azalt\u0131n.<\/li>\n
- Prompt M\u00fchendisli\u011fi:<\/strong> LLM'e verilen prompt'lar\u0131 optimize ederek daha k\u0131sa, daha do\u011fru ve daha h\u0131zl\u0131 yan\u0131tlar almay\u0131 hedefleyin.<\/li>\n<\/ul>\n
Ajan\u0131n Test Edilmesi ve Hata Ay\u0131klama<\/h4>\n\n- Birim Testleri:<\/strong> Her bir LangGraph d\u00fc\u011f\u00fcm\u00fcn\u00fcn (fonksiyonun) ba\u011f\u0131ms\u0131z olarak do\u011fru \u00e7al\u0131\u015ft\u0131\u011f\u0131ndan emin olmak i\u00e7in birim testleri yaz\u0131n.<\/li>\n
- Entegrasyon Testleri:<\/strong> Ajan\u0131n t\u00fcm ak\u0131\u015f\u0131n\u0131n (d\u00fc\u011f\u00fcmler aras\u0131 ge\u00e7i\u015fler dahil) beklendi\u011fi gibi \u00e7al\u0131\u015ft\u0131\u011f\u0131n\u0131 do\u011frulamak i\u00e7in entegrasyon testleri yap\u0131n.<\/li>\n
- U\u00e7tan Uca Testler:<\/strong> Ger\u00e7ek kullan\u0131c\u0131 sorgular\u0131yla ajan\u0131 test edin ve \u00e7\u0131kt\u0131lar\u0131n\u0131n (hem metin hem de grafik) do\u011fru ve faydal\u0131 oldu\u011fundan emin olun.<\/li>\n
- Hata Ay\u0131klama Ara\u00e7lar\u0131:<\/strong> LangGraph'\u0131n g\u00f6rselle\u015ftirme ara\u00e7lar\u0131n\u0131 veya LangSmith gibi izleme platformlar\u0131n\u0131 kullanarak ajan\u0131n ak\u0131\u015f\u0131n\u0131 ve hata noktalar\u0131n\u0131 tespit edin.<\/li>\n<\/ul>\n
\u00dcretime Ge\u00e7i\u015f ve \u00d6l\u00e7eklendirme<\/h4>\n\n- API Entegrasyonu:<\/strong> LangGraph ajan\u0131n\u0131z\u0131 bir web API'si (\u00f6rne\u011fin Flask, FastAPI) arkas\u0131na koyarak \u00f6n u\u00e7 uygulaman\u0131z\u0131n (CopilotKit) ona eri\u015fmesini sa\u011flay\u0131n.<\/li>\n
- Konteynerle\u015ftirme:<\/strong> Docker gibi ara\u00e7larla ajan\u0131 konteynerle\u015ftirerek farkl\u0131 ortamlarda tutarl\u0131 bir \u015fekilde \u00e7al\u0131\u015fmas\u0131n\u0131 sa\u011flay\u0131n.<\/li>\n
- Bulut Da\u011f\u0131t\u0131m\u0131:<\/strong> Ajan\u0131n\u0131z\u0131 ve API'nizi AWS, Azure veya Google Cloud gibi bulut platformlar\u0131na da\u011f\u0131t\u0131n. Sunucusuz fonksiyonlar (Lambda, Azure Functions) veya Kubernetes k\u00fcmeleri \u00f6l\u00e7eklenebilirlik i\u00e7in iyi se\u00e7enekler olabilir.<\/li>\n
- G\u00fcvenlik:<\/strong> API anahtarlar\u0131n\u0131z\u0131 ve hassas verileri g\u00fcvenli bir \u015fekilde saklay\u0131n. Kimlik do\u011frulama ve yetkilendirme mekanizmalar\u0131n\u0131 uygulay\u0131n.<\/li>\n
- \u0130zleme ve G\u00fcnl\u00fck Kayd\u0131:<\/strong> Ajan\u0131n performans\u0131n\u0131 ve hatalar\u0131n\u0131 izlemek i\u00e7in g\u00fcnl\u00fck kayd\u0131 ve izleme sistemleri kurun.<\/li>\n<\/ul>\n
Sonu\u00e7<\/h3>\n
Bu makalede, LangGraph'\u0131n g\u00fcc\u00fcn\u00fc kullanarak karma\u015f\u0131k ara\u015ft\u0131rma g\u00f6revlerini yerine getirebilen, ayn\u0131 zamanda CopilotKit ve Tako gibi MCP uygulamalar\u0131 arac\u0131l\u0131\u011f\u0131yla dinamik ve etkile\u015fimli grafiklerle zenginle\u015ftirilmi\u015f \u00e7\u0131kt\u0131lar sunan bir yapay zeka ajan\u0131 nas\u0131l olu\u015fturulaca\u011f\u0131n\u0131 detayl\u0131 bir \u015fekilde inceledik. LangGraph'\u0131n durum y\u00f6netimi ve d\u00f6ng\u00fc yetenekleri sayesinde ajan\u0131n ara\u015ft\u0131rma s\u00fcrecini esnek bir \u015fekilde tasarlayabilirken, CopilotKit ve Tako entegrasyonu ile analiz sonu\u00e7lar\u0131n\u0131 sadece metinle de\u011fil, g\u00f6rsel olarak da etkili bir \u015fekilde sunma kapasitesi kazand\u0131k. Bu entegre yakla\u015f\u0131m, kullan\u0131c\u0131lar\u0131n karma\u015f\u0131k verileri daha h\u0131zl\u0131 anlamas\u0131na, \u00f6nemli i\u00e7g\u00f6r\u00fcleri kolayca ke\u015ffetmesine ve nihayetinde daha bilin\u00e7li kararlar almas\u0131na olanak tan\u0131r. Yapay zeka destekli ara\u015ft\u0131rma ve veri g\u00f6rselle\u015ftirmenin birle\u015fimi, gelecekteki bilgiye eri\u015fim ve analiz y\u00f6ntemlerinin temelini olu\u015fturacakt\u0131r.<\/p>\n
\n","protected":false},"excerpt":{"rendered":"G\u00fcn\u00fcm\u00fcz\u00fcn h\u0131zla geli\u015fen teknoloji d\u00fcnyas\u0131nda, b\u00fcy\u00fck dil modellerinin (LLM) g\u00fcc\u00fcn\u00fc kullanarak karma\u015f\u0131k ara\u015ft\u0131rma g\u00f6revlerini otomatikle\u015ftir…","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"csco_page_header_type":"","csco_page_load_nextpost":"","csco_page_subscribe_form":"","csco_page_contact_form":"","footnotes":""},"categories":[1],"tags":[],"class_list":{"0":"post-38603","1":"post","2":"type-post","3":"status-publish","4":"format-standard","6":"category-genel","7":"cs-entry","8":"cs-video-wrap"},"yoast_head":"\nLangGraph Ara\u015ft\u0131rma Ajans\u0131 Olu\u015fturma: Dinamik Grafikleri MCP Uygulamalar\u0131 (CopilotKit & Tako) ile Entegre Etme - Kodlar\u0131n Gizemli D\u00fcnyas\u0131<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/fatihsoysal.com\/blog\/langgraph-arastirma-ajansi-olusturma-dinamik-grafikleri-mcp-uygulamalari-copilotkit-tako-ile-entegre-etme\/\" \/>\n<meta property=\"og:locale\" content=\"tr_TR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"LangGraph Ara\u015ft\u0131rma Ajans\u0131 Olu\u015fturma: Dinamik Grafikleri MCP Uygulamalar\u0131 (CopilotKit & Tako) ile Entegre Etme\" \/>\n<meta property=\"og:description\" content=\"G\u00fcn\u00fcm\u00fcz\u00fcn h\u0131zla geli\u015fen teknoloji d\u00fcnyas\u0131nda, b\u00fcy\u00fck dil modellerinin (LLM) g\u00fcc\u00fcn\u00fc kullanarak karma\u015f\u0131k ara\u015ft\u0131rma g\u00f6revlerini otomatikle\u015ftir...\" \/>\n<meta property=\"og:url\" content=\"https:\/\/fatihsoysal.com\/blog\/langgraph-arastirma-ajansi-olusturma-dinamik-grafikleri-mcp-uygulamalari-copilotkit-tako-ile-entegre-etme\/\" \/>\n<meta property=\"og:site_name\" content=\"Kodlar\u0131n Gizemli D\u00fcnyas\u0131\" \/>\n<meta property=\"article:published_time\" content=\"2026-02-03T16:00:46+00:00\" \/>\n<meta name=\"author\" content=\"Fatih Soysal\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Yazan:\" \/>\n\t<meta name=\"twitter:data1\" content=\"Fatih Soysal\" \/>\n\t<meta name=\"twitter:label2\" content=\"Tahmini okuma s\u00fcresi\" \/>\n\t<meta name=\"twitter:data2\" content=\"18 dakika\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/langgraph-arastirma-ajansi-olusturma-dinamik-grafikleri-mcp-uygulamalari-copilotkit-tako-ile-entegre-etme\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/langgraph-arastirma-ajansi-olusturma-dinamik-grafikleri-mcp-uygulamalari-copilotkit-tako-ile-entegre-etme\/\"},\"author\":{\"name\":\"Fatih Soysal\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1\"},\"headline\":\"LangGraph Ara\u015ft\u0131rma Ajans\u0131 Olu\u015fturma: Dinamik Grafikleri MCP Uygulamalar\u0131 (CopilotKit & Tako) ile Entegre Etme\",\"datePublished\":\"2026-02-03T16:00:46+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/langgraph-arastirma-ajansi-olusturma-dinamik-grafikleri-mcp-uygulamalari-copilotkit-tako-ile-entegre-etme\/\"},\"wordCount\":2761,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1\"},\"inLanguage\":\"tr\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\/\/fatihsoysal.com\/blog\/langgraph-arastirma-ajansi-olusturma-dinamik-grafikleri-mcp-uygulamalari-copilotkit-tako-ile-entegre-etme\/#respond\"]}],\"copyrightYear\":\"2026\",\"copyrightHolder\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#organization\"}},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/langgraph-arastirma-ajansi-olusturma-dinamik-grafikleri-mcp-uygulamalari-copilotkit-tako-ile-entegre-etme\/\",\"url\":\"https:\/\/fatihsoysal.com\/blog\/langgraph-arastirma-ajansi-olusturma-dinamik-grafikleri-mcp-uygulamalari-copilotkit-tako-ile-entegre-etme\/\",\"name\":\"LangGraph Ara\u015ft\u0131rma Ajans\u0131 Olu\u015fturma: Dinamik Grafikleri MCP Uygulamalar\u0131 (CopilotKit & Tako) ile Entegre Etme - Kodlar\u0131n Gizemli D\u00fcnyas\u0131\",\"isPartOf\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#website\"},\"datePublished\":\"2026-02-03T16:00:46+00:00\",\"breadcrumb\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/langgraph-arastirma-ajansi-olusturma-dinamik-grafikleri-mcp-uygulamalari-copilotkit-tako-ile-entegre-etme\/#breadcrumb\"},\"inLanguage\":\"tr\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/fatihsoysal.com\/blog\/langgraph-arastirma-ajansi-olusturma-dinamik-grafikleri-mcp-uygulamalari-copilotkit-tako-ile-entegre-etme\/\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/langgraph-arastirma-ajansi-olusturma-dinamik-grafikleri-mcp-uygulamalari-copilotkit-tako-ile-entegre-etme\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Anasayfa\",\"item\":\"https:\/\/fatihsoysal.com\/blog\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"LangGraph Ara\u015ft\u0131rma Ajans\u0131 Olu\u015fturma: Dinamik Grafikleri MCP Uygulamalar\u0131 (CopilotKit & Tako) ile Entegre Etme\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#website\",\"url\":\"https:\/\/fatihsoysal.com\/blog\/\",\"name\":\"Fatihsoysal.com\",\"description\":\"Blog - Yaz\u0131l\u0131m D\u00fcnyas\u0131 Tecr\u00fcbelerim\",\"publisher\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/fatihsoysal.com\/blog\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"tr\"},{\"@type\":[\"Person\",\"Organization\"],\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1\",\"name\":\"Fatih Soysal\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"tr\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/fatihsoysal.com\/blog\/wp-content\/uploads\/2024\/04\/cropped-replicate-prediction-3kgg1hgjn5rgp0cf0p5tr0jw7w-1.png\",\"contentUrl\":\"https:\/\/fatihsoysal.com\/blog\/wp-content\/uploads\/2024\/04\/cropped-replicate-prediction-3kgg1hgjn5rgp0cf0p5tr0jw7w-1.png\",\"width\":512,\"height\":512,\"caption\":\"Fatih Soysal\"},\"logo\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/image\/\"},\"description\":\"Kullan\u0131m ve kodlama m\u00fckemmeliyetini odak alan uygulamalar olu\u015fturma deneyimine sahip, profesyonel olarak 15+ y\u0131l \u00fczeri deneyime sahip bir yaz\u0131l\u0131m m\u00fchendisi.\",\"url\":\"https:\/\/fatihsoysal.com\/blog\/author\/fatihsoysal\/\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"LangGraph Ara\u015ft\u0131rma Ajans\u0131 Olu\u015fturma: Dinamik Grafikleri MCP Uygulamalar\u0131 (CopilotKit & Tako) ile Entegre Etme - Kodlar\u0131n Gizemli D\u00fcnyas\u0131","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/fatihsoysal.com\/blog\/langgraph-arastirma-ajansi-olusturma-dinamik-grafikleri-mcp-uygulamalari-copilotkit-tako-ile-entegre-etme\/","og_locale":"tr_TR","og_type":"article","og_title":"LangGraph Ara\u015ft\u0131rma Ajans\u0131 Olu\u015fturma: Dinamik Grafikleri MCP Uygulamalar\u0131 (CopilotKit & Tako) ile Entegre Etme","og_description":"G\u00fcn\u00fcm\u00fcz\u00fcn h\u0131zla geli\u015fen teknoloji d\u00fcnyas\u0131nda, b\u00fcy\u00fck dil modellerinin (LLM) g\u00fcc\u00fcn\u00fc kullanarak karma\u015f\u0131k ara\u015ft\u0131rma g\u00f6revlerini otomatikle\u015ftir...","og_url":"https:\/\/fatihsoysal.com\/blog\/langgraph-arastirma-ajansi-olusturma-dinamik-grafikleri-mcp-uygulamalari-copilotkit-tako-ile-entegre-etme\/","og_site_name":"Kodlar\u0131n Gizemli D\u00fcnyas\u0131","article_published_time":"2026-02-03T16:00:46+00:00","author":"Fatih Soysal","twitter_card":"summary_large_image","twitter_misc":{"Yazan:":"Fatih Soysal","Tahmini okuma s\u00fcresi":"18 dakika"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/fatihsoysal.com\/blog\/langgraph-arastirma-ajansi-olusturma-dinamik-grafikleri-mcp-uygulamalari-copilotkit-tako-ile-entegre-etme\/#article","isPartOf":{"@id":"https:\/\/fatihsoysal.com\/blog\/langgraph-arastirma-ajansi-olusturma-dinamik-grafikleri-mcp-uygulamalari-copilotkit-tako-ile-entegre-etme\/"},"author":{"name":"Fatih Soysal","@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1"},"headline":"LangGraph Ara\u015ft\u0131rma Ajans\u0131 Olu\u015fturma: Dinamik Grafikleri MCP Uygulamalar\u0131 (CopilotKit & Tako) ile Entegre Etme","datePublished":"2026-02-03T16:00:46+00:00","mainEntityOfPage":{"@id":"https:\/\/fatihsoysal.com\/blog\/langgraph-arastirma-ajansi-olusturma-dinamik-grafikleri-mcp-uygulamalari-copilotkit-tako-ile-entegre-etme\/"},"wordCount":2761,"commentCount":0,"publisher":{"@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1"},"inLanguage":"tr","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/fatihsoysal.com\/blog\/langgraph-arastirma-ajansi-olusturma-dinamik-grafikleri-mcp-uygulamalari-copilotkit-tako-ile-entegre-etme\/#respond"]}],"copyrightYear":"2026","copyrightHolder":{"@id":"https:\/\/fatihsoysal.com\/blog\/#organization"}},{"@type":"WebPage","@id":"https:\/\/fatihsoysal.com\/blog\/langgraph-arastirma-ajansi-olusturma-dinamik-grafikleri-mcp-uygulamalari-copilotkit-tako-ile-entegre-etme\/","url":"https:\/\/fatihsoysal.com\/blog\/langgraph-arastirma-ajansi-olusturma-dinamik-grafikleri-mcp-uygulamalari-copilotkit-tako-ile-entegre-etme\/","name":"LangGraph Ara\u015ft\u0131rma Ajans\u0131 Olu\u015fturma: Dinamik Grafikleri MCP Uygulamalar\u0131 (CopilotKit & Tako) ile Entegre Etme - Kodlar\u0131n Gizemli D\u00fcnyas\u0131","isPartOf":{"@id":"https:\/\/fatihsoysal.com\/blog\/#website"},"datePublished":"2026-02-03T16:00:46+00:00","breadcrumb":{"@id":"https:\/\/fatihsoysal.com\/blog\/langgraph-arastirma-ajansi-olusturma-dinamik-grafikleri-mcp-uygulamalari-copilotkit-tako-ile-entegre-etme\/#breadcrumb"},"inLanguage":"tr","potentialAction":[{"@type":"ReadAction","target":["https:\/\/fatihsoysal.com\/blog\/langgraph-arastirma-ajansi-olusturma-dinamik-grafikleri-mcp-uygulamalari-copilotkit-tako-ile-entegre-etme\/"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/fatihsoysal.com\/blog\/langgraph-arastirma-ajansi-olusturma-dinamik-grafikleri-mcp-uygulamalari-copilotkit-tako-ile-entegre-etme\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Anasayfa","item":"https:\/\/fatihsoysal.com\/blog\/"},{"@type":"ListItem","position":2,"name":"LangGraph Ara\u015ft\u0131rma Ajans\u0131 Olu\u015fturma: Dinamik Grafikleri MCP Uygulamalar\u0131 (CopilotKit & Tako) ile Entegre Etme"}]},{"@type":"WebSite","@id":"https:\/\/fatihsoysal.com\/blog\/#website","url":"https:\/\/fatihsoysal.com\/blog\/","name":"Fatihsoysal.com","description":"Blog - Yaz\u0131l\u0131m D\u00fcnyas\u0131 Tecr\u00fcbelerim","publisher":{"@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/fatihsoysal.com\/blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"tr"},{"@type":["Person","Organization"],"@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1","name":"Fatih Soysal","image":{"@type":"ImageObject","inLanguage":"tr","@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/image\/","url":"https:\/\/fatihsoysal.com\/blog\/wp-content\/uploads\/2024\/04\/cropped-replicate-prediction-3kgg1hgjn5rgp0cf0p5tr0jw7w-1.png","contentUrl":"https:\/\/fatihsoysal.com\/blog\/wp-content\/uploads\/2024\/04\/cropped-replicate-prediction-3kgg1hgjn5rgp0cf0p5tr0jw7w-1.png","width":512,"height":512,"caption":"Fatih Soysal"},"logo":{"@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/image\/"},"description":"Kullan\u0131m ve kodlama m\u00fckemmeliyetini odak alan uygulamalar olu\u015fturma deneyimine sahip, profesyonel olarak 15+ y\u0131l \u00fczeri deneyime sahip bir yaz\u0131l\u0131m m\u00fchendisi.","url":"https:\/\/fatihsoysal.com\/blog\/author\/fatihsoysal\/"}]}},"yoast_meta":{"yoast_wpseo_title":"","yoast_wpseo_metadesc":"","yoast_wpseo_canonical":""},"amp_enabled":true,"_links":{"self":[{"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/posts\/38603","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/comments?post=38603"}],"version-history":[{"count":0,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/posts\/38603\/revisions"}],"wp:attachment":[{"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/media?parent=38603"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/categories?post=38603"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/tags?post=38603"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}