Auto-respond to Gmail Enquiries with GPT-4o & AI Agent

The workflow JSON

Copy or download the full n8n JSON below. Paste it into a new n8n workflow, add your credentials, activate. Full import guide →

{
  "id": "uQTqLH5At9PZOy2g",
  "meta": {
    "templateCredsSetupCompleted": true
  },
  "name": "Auto-Respond to Gmail Enquiries using GPT-4o, Dumpling AI & LangChain Agent",
  "tags": [],
  "nodes": [
    {
      "id": "149c7dd9-9f5f-4c5b-8342-7c907eb52402",
      "name": "Watch Gmail for New Incoming Emails",
      "type": "n8n-nodes-base.gmailTrigger",
      "position": [
        -2300,
        -340
      ],
      "parameters": {
        "filters": {},
        "pollTimes": {
          "item": [
            {
              "mode": "everyMinute"
            }
          ]
        }
      },
      "credentials": {
        "gmailOAuth2": {
          "name": "<your credential>"
        }
      },
      "typeVersion": 1.2
    },
    {
      "id": "8a0de384-9028-47e3-86e5-b5bf6476ae0c",
      "name": "Classify Email Type with GPT-4o",
      "type": "@n8n/n8n-nodes-langchain.openAi",
      "position": [
        -2080,
        -340
      ],
      "parameters": {
        "modelId": {
          "__rl": true,
          "mode": "list",
          "value": "gpt-4o",
          "cachedResultName": "GPT-4O"
        },
        "options": {},
        "messages": {
          "values": [
            {
              "content": "=Classify the following email content. Determine if it is an enquiry.\nIf it is an enquiry, return only this word: enquiry\nIf it is not an enquiry, return only this word: false\nDo not explain or add any other text. Only return the result.\nHere is the email body: {{ $json.snippet }}"
            }
          ]
        }
      },
      "credentials": {
        "openAiApi": {
          "name": "<your credential>"
        }
      },
      "typeVersion": 1.8
    },
    {
      "id": "57756cd3-9009-4189-8cba-e2edba87554f",
      "name": " Only Proceed if Email is an Enquiry",
      "type": "n8n-nodes-base.filter",
      "position": [
        -1704,
        -340
      ],
      "parameters": {
        "options": {},
        "conditions": {
          "options": {
            "version": 2,
            "leftValue": "",
            "caseSensitive": true,
            "typeValidation": "strict"
          },
          "combinator": "and",
          "conditions": [
            {
              "id": "e6019006-0980-4cd6-8e5d-3f618fbca13a",
              "operator": {
                "name": "filter.operator.equals",
                "type": "string",
                "operation": "equals"
              },
              "leftValue": "={{ $json.message.content }}",
              "rightValue": "enquiry"
            }
          ]
        }
      },
      "typeVersion": 2.2
    },
    {
      "id": "64e851b8-7ff1-4b4c-9828-7c764594f1ea",
      "name": "GPT-4o Chat Model",
      "type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
      "position": [
        -1484,
        -120
      ],
      "parameters": {
        "model": {
          "__rl": true,
          "mode": "list",
          "value": "gpt-4o-mini"
        },
        "options": {}
      },
      "credentials": {
        "openAiApi": {
          "name": "<your credential>"
        }
      },
      "typeVersion": 1.2
    },
    {
      "id": "2957c6ad-5e7b-4692-a96f-5c4fb5b8203c",
      "name": "Memory Buffer (Past 10 Interactions)",
      "type": "@n8n/n8n-nodes-langchain.memoryBufferWindow",
      "position": [
        -1364,
        -120
      ],
      "parameters": {
        "contextWindowLength": 10
      },
      "typeVersion": 1.3
    },
    {
      "id": "9e8617f1-cd89-4de5-9d44-ded155779745",
      "name": "Dumpling AI Agent \u2013 Search for Relevant Info",
      "type": "@n8n/n8n-nodes-langchain.toolHttpRequest",
      "position": [
        -1244,
        -120
      ],
      "parameters": {
        "url": "https://app.dumplingai.com/api/v1/agents/generate-completion",
        "method": "POST",
        "jsonBody": "={\n  \"messages\": [\n    {\n      \"role\": \"user\",\n      \"content\":\"{{ $('Watch Gmail for New Incoming Emails').item.json.snippet }}\"\n    }\n  ],\n  \"agentId\": \"a88a9b6c-1578-4da2-800b-561327367713\",\n  \"parseJson\": \"True\"\n  }",
        "sendBody": true,
        "specifyBody": "json",
        "authentication": "genericCredentialType",
        "genericAuthType": "httpHeaderAuth"
      },
      "credentials": {
        "httpHeaderAuth": {
          "name": "<your credential>"
        }
      },
      "typeVersion": 1.1
    },
    {
      "id": "da474406-a6e3-427d-b21b-3e7e00ebe910",
      "name": "Send Email Response via Gmail",
      "type": "n8n-nodes-base.gmailTool",
      "position": [
        -1124,
        -120
      ],
      "parameters": {
        "sendTo": "",
        "message": "={{ /*n8n-auto-generated-fromAI-override*/ $fromAI('Message', ``, 'string') }}",
        "options": {},
        "subject": "={{ /*n8n-auto-generated-fromAI-override*/ $fromAI('Subject', ``, 'string') }}"
      },
      "credentials": {
        "gmailOAuth2": {
          "name": "<your credential>"
        }
      },
      "typeVersion": 2.1
    },
    {
      "id": "03882650-9272-43c3-b1bc-15adae456881",
      "name": "LangChain Agent Handles Reply Logic",
      "type": "@n8n/n8n-nodes-langchain.agent",
      "position": [
        -1392,
        -340
      ],
      "parameters": {
        "text": "={{ $('Watch Gmail for New Incoming Emails').item.json.snippet }}",
        "options": {
          "systemMessage": "You are a helpful assistant, use the search for information tool to search for users information.\n\nuse Gmail tool to send email"
        },
        "promptType": "define"
      },
      "typeVersion": 1.8
    },
    {
      "id": "52bb3b65-0b7f-4206-87c9-1f91997a44dd",
      "name": "Sticky Note",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -1500,
        -560
      ],
      "parameters": {
        "width": 480,
        "height": 240,
        "content": "### \u2709\ufe0f Sticky Note 2: Website Scraping, Email Generation, and Sending\n\nFor leads with complete data, Dumpling AI scrapes the content of the lead\u2019s company website. This content, combined with scraped details, is structured into a prompt using the Set node. GPT-4o processes the prompt and generates a personalized cold email. The email is then sent to the lead using Gmail. This automates targeted cold outreach using scraped insights and AI-generated messaging.\n"
      },
      "typeVersion": 1
    },
    {
      "id": "3a6df2e9-1d1a-4cab-889c-c323f876b98f",
      "name": "Sticky Note1",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -2320,
        -540
      ],
      "parameters": {
        "width": 540,
        "height": 220,
        "content": "### Sticky Note 1: Apollo Link Sourcing & Contact Scraping\n\nThis section begins the workflow manually. The Apollo lead URLs are pulled from a Google Sheet. Each link is passed to Apify which scrapes key lead details such as name, email, company website, and other contact data. After scraping, a filter checks that both an email and website were found for the lead before moving forward. This ensures only complete lead data is used for outreach.\n\n"
      },
      "typeVersion": 1
    }
  ],
  "active": false,
  "settings": {
    "executionOrder": "v1"
  },
  "versionId": "0f1d6a0b-eb13-4e1a-be9e-3f22eb17e66e",
  "connections": {
    "GPT-4o Chat Model": {
      "ai_languageModel": [
        [
          {
            "node": "LangChain Agent Handles Reply Logic",
            "type": "ai_languageModel",
            "index": 0
          }
        ]
      ]
    },
    "Send Email Response via Gmail": {
      "ai_tool": [
        [
          {
            "node": "LangChain Agent Handles Reply Logic",
            "type": "ai_tool",
            "index": 0
          }
        ]
      ]
    },
    "Classify Email Type with GPT-4o": {
      "main": [
        [
          {
            "node": " Only Proceed if Email is an Enquiry",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "LangChain Agent Handles Reply Logic": {
      "main": [
        []
      ]
    },
    "Watch Gmail for New Incoming Emails": {
      "main": [
        [
          {
            "node": "Classify Email Type with GPT-4o",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    " Only Proceed if Email is an Enquiry": {
      "main": [
        [
          {
            "node": "LangChain Agent Handles Reply Logic",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Memory Buffer (Past 10 Interactions)": {
      "ai_memory": [
        [
          {
            "node": "LangChain Agent Handles Reply Logic",
            "type": "ai_memory",
            "index": 0
          }
        ]
      ]
    },
    "Dumpling AI Agent \u2013 Search for Relevant Info": {
      "ai_tool": [
        [
          {
            "node": "LangChain Agent Handles Reply Logic",
            "type": "ai_tool",
            "index": 0
          }
        ]
      ]
    }
  }
}

Credentials you'll need

Each integration node will prompt for credentials when you import. We strip credential IDs before publishing — you'll add your own.

• gmailOAuth2
• httpHeaderAuth
• openAiApi

About this workflow

This workflow is perfect for customer support teams, sales departments, or solopreneurs who receive frequent email enquiries and want to automate the initial response process using AI. If you spend too much time answering similar questions, this system helps respond faster and…

Source: https://n8n.io/workflows/4057/ — original creator credit. Request a take-down →

More AI & RAG workflows → · Browse all categories →

Related workflows

Workflows that share integrations, category, or trigger type with this one. All free to copy and import.