8
n8n 한국어amn8n.com

Google Drive와 Pinecone 벡터 데이터베이스를 사용하여 AI 기반 WhatsApp 로봇 생성

고급

이것은Support, AI분야의자동화 워크플로우로, 23개의 노드를 포함합니다.주로 If, Set, Code, Webhook, GoogleDrive 등의 노드를 사용하며인공지능 기술을 결합하여 스마트 자동화를 구현합니다. Google Drive와 Pinecone 벡터 데이터베이스를 사용하여 AI로운 WhatsApp 로봇을 만듭니다.

사전 요구사항
  • HTTP Webhook 엔드포인트(n8n이 자동으로 생성)
  • Google Drive API 인증 정보
  • OpenAI API Key
  • Pinecone API Key

카테고리

워크플로우 미리보기
노드 연결 관계를 시각적으로 표시하며, 확대/축소 및 이동을 지원합니다
워크플로우 내보내기
다음 JSON 구성을 복사하여 n8n에 가져오면 이 워크플로우를 사용할 수 있습니다
{
  "meta": {
    "instanceId": "e42ff03bc515cd7f9edc081a89072ca1985132e0d4a1cdabe6c26855a252768f",
    "templateId": "1960",
    "templateCredsSetupCompleted": true
  },
  "nodes": [
    {
      "id": "1f2bb917-6d65-4cfa-9474-fc3b19a8c3bd",
      "name": "Sticky Note",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        60,
        200
      ],
      "parameters": {
        "color": 7,
        "width": 1218,
        "height": 627,
        "content": "## Load Information from a Google Drive File into a Pinecone Vector Database\nFetch the file from Google Drive, split it into chunks, and insert the chunks into a Pinecone index."
      },
      "typeVersion": 1
    },
    {
      "id": "ab577f4d-8906-4e0c-bc62-e8a4b2610551",
      "name": "Sticky Note2",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -500,
        260
      ],
      "parameters": {
        "width": 480,
        "height": 145,
        "content": "## Step 1: Connect Your Google Drive  \nFollow the instructions in the official n8n docs to set up your Google Drive credentials: \n👉 [Google Drive Credentials Guide](https://docs.n8n.io/integrations/builtin/credentials/google/)"
      },
      "typeVersion": 1
    },
    {
      "id": "742beb54-8b89-49a3-afe5-fd7e73b37044",
      "name": "Default Data Loader",
      "type": "@n8n/n8n-nodes-langchain.documentDefaultDataLoader",
      "position": [
        720,
        500
      ],
      "parameters": {
        "options": {},
        "dataType": "binary"
      },
      "typeVersion": 1
    },
    {
      "id": "f75e31e9-f752-45d1-bc44-75097ec85ce6",
      "name": "Recursive Character Text Splitter",
      "type": "@n8n/n8n-nodes-langchain.textSplitterRecursiveCharacterTextSplitter",
      "position": [
        800,
        620
      ],
      "parameters": {
        "options": {},
        "chunkSize": 3000,
        "chunkOverlap": 200
      },
      "typeVersion": 1
    },
    {
      "id": "bac883c8-4c1f-466b-b20f-d0fdf6acfc42",
      "name": "Embeddings OpenAI",
      "type": "@n8n/n8n-nodes-langchain.embeddingsOpenAi",
      "position": [
        600,
        500
      ],
      "parameters": {
        "options": {}
      },
      "credentials": {
        "openAiApi": {
          "id": "wsxLB7on237QOp5h",
          "name": "Open AI credentials"
        }
      },
      "typeVersion": 1.2
    },
    {
      "id": "430f667b-9d23-4bb9-b87d-5d9e8ae84df3",
      "name": "Simple Memory",
      "type": "@n8n/n8n-nodes-langchain.memoryBufferWindow",
      "position": [
        1500,
        1180
      ],
      "parameters": {
        "sessionKey": "={{ $json.fromNumber }}",
        "sessionIdType": "customKey",
        "contextWindowLength": 20
      },
      "typeVersion": 1.3
    },
    {
      "id": "1a2e6717-a220-4f39-8ae9-1921cbff5715",
      "name": "OpenAI Chat Model",
      "type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
      "position": [
        1260,
        1200
      ],
      "parameters": {
        "model": {
          "__rl": true,
          "mode": "list",
          "value": "gpt-4o-mini"
        },
        "options": {}
      },
      "credentials": {
        "openAiApi": {
          "id": "wsxLB7on237QOp5h",
          "name": "Open AI credentials"
        }
      },
      "typeVersion": 1.2
    },
    {
      "id": "7e03e08d-0b61-4054-aa14-8f839d7f0565",
      "name": "Question & Answer",
      "type": "@n8n/n8n-nodes-langchain.agent",
      "position": [
        1380,
        960
      ],
      "parameters": {
        "text": "=Sender Name: {{ $json.fromName }}\nMessage: {{ $json.textBody }}",
        "options": {
          "systemMessage": "=Agent Name: Van\nIdentity:\nYou are a friendly AI agent named Van. You are professional digital sales agent for GrayBox, a company specializing in Structural Insulated Panels (SIP) construction. Van immediately identifies as a digital assistant and exists to serve prospective and current customers of GrayBox with accurate, helpful, and engaging information that supports decision-making and promotes trust.\n\nYour Personality & Style\nTone: Clear, concise, professional—but always friendly and approachable. Talk in first person representing the company.\n\n\nIf user send link don't mention that you cannot read the link.  Explain who you are, and ask questions to engage the conversation..  Your goal is to sell our solution.  \n\nResponse Style:\n\n\nGreet, answer directly, then offer further help.\nAddress people by their names\n\n\n\nAvoid fluff and unnecessary formalities (unless it's the first message).\n\n\nMaintain consistent use of brand tone: Speed, Quality, Efficiency, Value.\n\n\nBrand Voice Keywords: Speed, Energy Efficiency, Durability, Mold-Free, Earthquake-Resistant, Friendly, Human.\n\n\n\n🔍 Scope of the Agent (What Van Can Do)\nAnswer questions about pricing, materials, SIP benefits, timeline, logistics, and project experience.\n\n\nExplain how GrayBox SIP construction works (in simple, non-technical language).\n\n\nShare relevant visual or catalog links.\n\n\nGuide users to:\n\n\nSchedule a 15-min discovery call: https://calendar.app.google/gJYWhw3LCJBPywGW6\n\n\nVisit projects like Nuanu City or Luna Beach Club.\n\n\nSubmit project materials via form: https://forms.gle/5AddUQbqveFPsTDQ9\nView portfolio: https://graybox.site/graybox-0-0 \n\n🧠 Core Behaviors & Instructions\nAlways say cost starts at 3 million IDR/m² for core and shell.\n\n\nPromote:\n\n\n1-month build time (from materials delivery).\n\n\nCost-effectiveness vs. traditional methods (20–30% cheaper, 4x faster).\n\n\nEnergy savings and safety (mold-free, earthquake-resistant).\n\n\nWhen users upload a file/image: respond with a friendly message and prompt them to fill the quotation form.\n\n\nIf asked about pictures, share: https://graybox.site/graybox-0-0\n\n\nFor meetings or calls: Share this link → https://calendar.app.google/rDzpGdMb3WmGsLFz6\n\n\nFor human contact or visit coordination: Provide Andrea's WhatsApp: +1 650 866 5016\n\n\nIf user says “OK, I’ll visit,” connect directly to Andrea.\n\n\n\n🌍 Contact & Location\nOffice: Uluwatu, Bali – https://maps.app.goo.gl/nT7VDTtLQYSztN7s9\nFactory: Lombok\n\n\nWhatsApp (Andrea): +1 650 866 5016 (only if meeting/human contact requested)\n\n🚀 First-Time Message (Greeting Script)\nDiscover GrayBox: Your dream home starts here!\n We deliver core and shell construction in under a month for developers, builders, and DIY home enthusiasts. Save time and money while enjoying a mold-free, energy-efficient, and earthquake-resistant home. All for just 3 million IDR per m²!\n Build smarter with GrayBox today. \n\nNEVER MAKEUP INFORMATION. \nUSE  {{ $json.context }} TO ANSWER QUESTIONS"
        },
        "promptType": "define"
      },
      "typeVersion": 1.8
    },
    {
      "id": "56d150ba-9b98-4dbb-9f74-5379344d2257",
      "name": "Whatsapp Webhook 응답",
      "type": "n8n-nodes-base.respondToWebhook",
      "position": [
        1720,
        960
      ],
      "parameters": {
        "options": {}
      },
      "typeVersion": 1.2
    },
    {
      "id": "5d1d4427-bdab-49ef-acbc-084a2d69632f",
      "name": "청크 준비",
      "type": "n8n-nodes-base.code",
      "position": [
        1200,
        960
      ],
      "parameters": {
        "jsCode": "let out = \"\"\nfor (const i in $input.all()) {\n  let itemText = \"--- CHUNK \" + i + \" ---\\n\"\n  itemText += $input.all()[i].json.document.pageContent + \"\\n\"\n  itemText += \"\\n\"\n  out += itemText\n}\n\nreturn {\n  'context': out,\n  'textBody': $('Listen to Whatsapp webhook').first().json.body.textBody,\n  'fromNumber': $('Listen to Whatsapp webhook').first().json.body.fromNumber,\n  'fromName': $('Listen to Whatsapp webhook').first().json.body.fromName\n};"
      },
      "typeVersion": 2
    },
    {
      "id": "eba8eaca-915f-4257-901e-c6f56ddf8f0b",
      "name": "쿼리와 일치하는 상위 청크 가져오기",
      "type": "@n8n/n8n-nodes-langchain.vectorStorePinecone",
      "position": [
        840,
        960
      ],
      "parameters": {
        "mode": "load",
        "topK": "={{ $json.chunks }}",
        "prompt": "={{ $json.body.textBody }}",
        "options": {},
        "pineconeIndex": {
          "__rl": true,
          "mode": "list",
          "value": "3vansales",
          "cachedResultName": "3vansales"
        }
      },
      "credentials": {
        "pineconeApi": {
          "id": "2ZUBIWHNyYqKmASZ",
          "name": "PineconeApi account 2"
        }
      },
      "typeVersion": 1
    },
    {
      "id": "afb0bb4f-8370-4a7a-be3c-8449f5b890d5",
      "name": "모델에 전송할 최대 청크 설정",
      "type": "n8n-nodes-base.set",
      "position": [
        620,
        960
      ],
      "parameters": {
        "options": {},
        "assignments": {
          "assignments": [
            {
              "id": "33f4addf-72f3-4618-a6ba-5b762257d723",
              "name": "chunks",
              "type": "number",
              "value": 4
            }
          ]
        },
        "includeOtherFields": true
      },
      "typeVersion": 3.4
    },
    {
      "id": "d9aa944d-594f-49ad-a40e-123aa42621e3",
      "name": "Embeddings OpenAI2",
      "type": "@n8n/n8n-nodes-langchain.embeddingsOpenAi",
      "position": [
        840,
        1180
      ],
      "parameters": {
        "options": {}
      },
      "credentials": {
        "openAiApi": {
          "id": "wsxLB7on237QOp5h",
          "name": "Open AI credentials"
        }
      },
      "typeVersion": 1.2
    },
    {
      "id": "15ff0895-9bb2-47d3-92fb-d0d8d1f4a981",
      "name": "개인 또는 그룹 메시지 확인",
      "type": "n8n-nodes-base.if",
      "position": [
        340,
        980
      ],
      "parameters": {
        "options": {},
        "conditions": {
          "options": {
            "version": 2,
            "leftValue": "",
            "caseSensitive": true,
            "typeValidation": "strict"
          },
          "combinator": "and",
          "conditions": [
            {
              "id": "5a946369-3f38-4dcb-b5d9-3d24024d08b0",
              "operator": {
                "name": "filter.operator.equals",
                "type": "string",
                "operation": "equals"
              },
              "leftValue": "={{ $json.body.threadType }}",
              "rightValue": "individual"
            }
          ]
        }
      },
      "typeVersion": 2.2
    },
    {
      "id": "32df5bff-1b09-4f1c-9344-7d9ae748d917",
      "name": "Whatsapp webhook 수신 대기",
      "type": "n8n-nodes-base.webhook",
      "position": [
        120,
        980
      ],
      "webhookId": "d1629cd7-6e05-413a-bac0-3c435d4e43b9",
      "parameters": {
        "path": "d1629cd7-6e05-413a-bac0-3c435d4e43b9",
        "options": {},
        "httpMethod": "POST",
        "responseMode": "responseNode"
      },
      "typeVersion": 2
    },
    {
      "id": "8b493fb6-8013-453f-8aa5-1ee9c9d06cd9",
      "name": "Sticky Note1",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        60,
        860
      ],
      "parameters": {
        "color": 7,
        "width": 2034,
        "height": 627,
        "content": "### Chat with Your Vector Database Using a Custom WhatsApp Webhook (with Group Support)\nThis workflow takes incoming WhatsApp messages, embeds the content, retrieves relevant chunks from your vector store, and passes them to the model to generate a response."
      },
      "typeVersion": 1
    },
    {
      "id": "d97920ad-6b36-4981-8b9d-9d470b5c769a",
      "name": "Google Drive에서 파일 다운로드",
      "type": "n8n-nodes-base.googleDrive",
      "position": [
        360,
        280
      ],
      "parameters": {
        "fileId": {
          "__rl": true,
          "mode": "id",
          "value": "={{ $json.id }}"
        },
        "options": {},
        "operation": "download",
        "authentication": "serviceAccount"
      },
      "credentials": {
        "googleApi": {
          "id": "10HrAmRHIIsSrT6B",
          "name": "Google Drive account"
        }
      },
      "typeVersion": 3
    },
    {
      "id": "0a6f8b88-9c62-4e3e-82cb-a7028bdcac45",
      "name": "Pinecone 벡터 저장소 인덱싱",
      "type": "@n8n/n8n-nodes-langchain.vectorStorePinecone",
      "position": [
        620,
        280
      ],
      "parameters": {
        "mode": "insert",
        "options": {
          "clearNamespace": true
        },
        "pineconeIndex": {
          "__rl": true,
          "mode": "list",
          "value": "3vansales",
          "cachedResultName": "3vansales"
        }
      },
      "credentials": {
        "pineconeApi": {
          "id": "2ZUBIWHNyYqKmASZ",
          "name": "PineconeApi account 2"
        }
      },
      "typeVersion": 1
    },
    {
      "id": "34f220a8-12d7-480f-9954-11d777359d27",
      "name": "파일 업데이트 여부 1분마다 확인",
      "type": "n8n-nodes-base.googleDriveTrigger",
      "position": [
        140,
        280
      ],
      "parameters": {
        "pollTimes": {
          "item": [
            {
              "mode": "everyMinute"
            }
          ]
        },
        "triggerOn": "specificFile",
        "fileToWatch": {
          "__rl": true,
          "mode": "list",
          "value": "1-AZrzOeeZFATUsY-xGD0m_oYkq4ekiSHn3e8Aof9pyU",
          "cachedResultUrl": "https://docs.google.com/document/d/1-AZrzOeeZFATUsY-xGD0m_oYkq4ekiSHn3e8Aof9pyU/edit?usp=drivesdk",
          "cachedResultName": "Knowladge Base FlatPack"
        },
        "authentication": "serviceAccount"
      },
      "credentials": {
        "googleApi": {
          "id": "10HrAmRHIIsSrT6B",
          "name": "Google Drive account"
        }
      },
      "typeVersion": 1
    },
    {
      "id": "f9016922-f6a4-417a-8586-bf0647f1e17b",
      "name": "Sticky Note3",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -500,
        440
      ],
      "parameters": {
        "width": 480,
        "height": 325,
        "content": "## Step 2: Set Up Your Pinecone Vector Database \n1. Log in to your Pinecone account and create an index: \n🚨 Dimensions: 1536\n2. In the workflow, make sure to select this index in **both Pinecone nodes**.\n3. Click \"Test Workflow\" to load your data into the vector database.\n4. Modify the connected document to fit your use case.\n\n**Note:** Once the bot is live, the vector database updates automatically every minute when changes are made. You can customize this update frequency if needed."
      },
      "typeVersion": 1
    },
    {
      "id": "638760a3-51da-4b88-ace5-73e67e6e6e98",
      "name": "Sticky Note4",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -500,
        800
      ],
      "parameters": {
        "color": 2,
        "width": 480,
        "height": 205,
        "content": "## Step 3: Request Access to the Custom WhatsApp Webhook\n1. Fill out this form to request access: \n👉 [Request Access](https://docs.google.com/forms/d/e/1FAIpQLSd-bW5tSJu_rRvJ4NmFrxXSAwaNbO7MbGJtUIS-mBA23B7BWQ/viewform?usp=dialog)\n2. You’ll receive a confirmation message on WhatsApp to verify your connection.\n3. Customize the **AI agent's system prompt** to match your use case."
      },
      "typeVersion": 1
    },
    {
      "id": "a9092d34-7406-4d7e-8fe4-ce392aae9c5a",
      "name": "Sticky Note5",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -500,
        1040
      ],
      "parameters": {
        "color": 2,
        "width": 480,
        "height": 345,
        "content": "## Step 4: Test the WhatsApp Integration\n#### One-on-One Testing\n1. Use a different WhatsApp number to send a message to the connected number.\n2. You should receive a reply from the AI based on:\n  ✅ Your system prompt\n  ✅ The content in your vector database\n#### Group Testing\n1. Add the bot to a WhatsApp group. \n2. The bot will only reply when tagged or addressed directly, to avoid responding to every message in the group"
      },
      "typeVersion": 1
    },
    {
      "id": "8b338d65-cd79-41d5-b4c4-e76534d8b19c",
      "name": "Sticky Note6",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -1160,
        260
      ],
      "parameters": {
        "width": 480,
        "height": 645,
        "content": "## 🔑 Key Benefits\n#### 1. No Meta Account Needed\n👉 Works with regular WhatsApp numbers — no Facebook Business setup or official API required.\n\n#### 2. Group Chat Support\n👉 Add the bot to any group. It only replies when tagged to avoid cluttering conversations.\n\n#### 3. Custom AI Responses\n👉 Connect your own documents (via Google Drive + Pinecone) to deliver context-aware replies.\n\n#### 4. Auto-Syncing Knowledge Base\n👉 Updates every minute when your source document changes — always fresh, always relevant.\n\n#### Use cases are endless\n✅ Customer support\n✅ Internal team assistant\n✅ Community group bot\n✅ Personal knowledge brain\n\nAnd so on "
      },
      "typeVersion": 1
    }
  ],
  "pinData": {},
  "connections": {
    "430f667b-9d23-4bb9-b87d-5d9e8ae84df3": {
      "ai_memory": [
        [
          {
            "node": "7e03e08d-0b61-4054-aa14-8f839d7f0565",
            "type": "ai_memory",
            "index": 0
          }
        ]
      ]
    },
    "5d1d4427-bdab-49ef-acbc-084a2d69632f": {
      "main": [
        [
          {
            "node": "7e03e08d-0b61-4054-aa14-8f839d7f0565",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "bac883c8-4c1f-466b-b20f-d0fdf6acfc42": {
      "ai_embedding": [
        [
          {
            "node": "0a6f8b88-9c62-4e3e-82cb-a7028bdcac45",
            "type": "ai_embedding",
            "index": 0
          }
        ]
      ]
    },
    "1a2e6717-a220-4f39-8ae9-1921cbff5715": {
      "ai_languageModel": [
        [
          {
            "node": "7e03e08d-0b61-4054-aa14-8f839d7f0565",
            "type": "ai_languageModel",
            "index": 0
          }
        ]
      ]
    },
    "7e03e08d-0b61-4054-aa14-8f839d7f0565": {
      "main": [
        [
          {
            "node": "56d150ba-9b98-4dbb-9f74-5379344d2257",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "d9aa944d-594f-49ad-a40e-123aa42621e3": {
      "ai_embedding": [
        [
          {
            "node": "eba8eaca-915f-4257-901e-c6f56ddf8f0b",
            "type": "ai_embedding",
            "index": 0
          }
        ]
      ]
    },
    "742beb54-8b89-49a3-afe5-fd7e73b37044": {
      "ai_document": [
        [
          {
            "node": "0a6f8b88-9c62-4e3e-82cb-a7028bdcac45",
            "type": "ai_document",
            "index": 0
          }
        ]
      ]
    },
    "32df5bff-1b09-4f1c-9344-7d9ae748d917": {
      "main": [
        [
          {
            "node": "15ff0895-9bb2-47d3-92fb-d0d8d1f4a981",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "eba8eaca-915f-4257-901e-c6f56ddf8f0b": {
      "main": [
        [
          {
            "node": "5d1d4427-bdab-49ef-acbc-084a2d69632f",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "afb0bb4f-8370-4a7a-be3c-8449f5b890d5": {
      "main": [
        [
          {
            "node": "eba8eaca-915f-4257-901e-c6f56ddf8f0b",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "f75e31e9-f752-45d1-bc44-75097ec85ce6": {
      "ai_textSplitter": [
        [
          {
            "node": "742beb54-8b89-49a3-afe5-fd7e73b37044",
            "type": "ai_textSplitter",
            "index": 0
          }
        ]
      ]
    },
    "d97920ad-6b36-4981-8b9d-9d470b5c769a": {
      "main": [
        [
          {
            "node": "0a6f8b88-9c62-4e3e-82cb-a7028bdcac45",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "34f220a8-12d7-480f-9954-11d777359d27": {
      "main": [
        [
          {
            "node": "d97920ad-6b36-4981-8b9d-9d470b5c769a",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "15ff0895-9bb2-47d3-92fb-d0d8d1f4a981": {
      "main": [
        [
          {
            "node": "afb0bb4f-8370-4a7a-be3c-8449f5b890d5",
            "type": "main",
            "index": 0
          }
        ]
      ]
    }
  }
}
자주 묻는 질문

이 워크플로우를 어떻게 사용하나요?

위의 JSON 구성 코드를 복사하여 n8n 인스턴스에서 새 워크플로우를 생성하고 "JSON에서 가져오기"를 선택한 후, 구성을 붙여넣고 필요에 따라 인증 설정을 수정하세요.

이 워크플로우는 어떤 시나리오에 적합한가요?

고급 - 지원, 인공지능

유료인가요?

이 워크플로우는 완전히 무료이며 직접 가져와 사용할 수 있습니다. 다만, 워크플로우에서 사용하는 타사 서비스(예: OpenAI API)는 사용자 직접 비용을 지불해야 할 수 있습니다.

워크플로우 정보
난이도
고급
노드 수23
카테고리2
노드 유형15
난이도 설명

고급 사용자를 위한 16+개 노드의 복잡한 워크플로우

외부 링크
n8n.io에서 보기

이 워크플로우 공유

카테고리

카테고리: 34