使用GPT-4o将自然语言转换为Google Sheets SQL查询

• •可能需要目标 API 的认证凭证
• •Google Sheets API 凭证
• •OpenAI API Key

{
  "meta": {
    "instanceId": "ad0113c344ee237399e44e9f11798b05baeb83a6196d514a9ae9d2ad71c3b5c9",
    "templateCredsSetupCompleted": true
  },
  "nodes": [
    {
      "id": "5a4c2536-9070-40b4-9db1-24e3b6f440fa",
      "name": "AI Agent3",
      "type": "@n8n/n8n-nodes-langchain.agent",
      "position": [
        224,
        1872
      ],
      "parameters": {
        "options": {
          "systemMessage": "# MVP System Prompt — Google Sheets Ask-Data\n\nRule, you must always use the column tool for every question and followup question to know the column names and column letters. \n\nwrite a sql query to send to google sheets based on the location of the google column on the google sheet. The output should look something like this. \n\nmust include google sheets entire url before the query. do not truncate. \"https://docs.google.com/spreadsheets/d/19aUQYZq02qHsCelO4eeV4sx_MTJJupC5qe0gDLQBtRA/edit?gid=365710158#gid=365710158\" \n\nexample: \nhttps://docs.google.com/spreadsheets/d/19aUQYZq02qHsCelO4eeV4sx_MTJJupC5qe0gDLQBtRA/gviz/tq?tqx=out:csv&gid=365710158&tq=SELECT E, SUM(H), SUM(F) GROUP BY E\n\nYou can only use column letters, no names\nouptut data like this. \n\n{\n\t\"query\": \"https://docs.google.com/spreadsheets/d/19aUQYZq02qHsCelO4eeV4sx_MTJJupC5qe0gDLQBtRA/gviz/tq?tqx=out:csv&gid=365710158&tq=SELECT E, SUM(H), SUM(F) GROUP BY E\"\n}"
        },
        "hasOutputParser": true
      },
      "typeVersion": 2.2
    },
    {
      "id": "766b4958-6e0b-4d99-99ce-8fe98a44b019",
      "name": "获取列信息2",
      "type": "n8n-nodes-base.googleSheetsTool",
      "position": [
        368,
        2160
      ],
      "parameters": {
        "options": {},
        "sheetName": {
          "__rl": true,
          "mode": "list",
          "value": 467321788,
          "cachedResultUrl": "https://docs.google.com/spreadsheets/d/19aUQYZq02qHsCelO4eeV4sx_MTJJupC5qe0gDLQBtRA/edit#gid=467321788",
          "cachedResultName": "Columns"
        },
        "documentId": {
          "__rl": true,
          "mode": "list",
          "value": "19aUQYZq02qHsCelO4eeV4sx_MTJJupC5qe0gDLQBtRA",
          "cachedResultUrl": "https://docs.google.com/spreadsheets/d/19aUQYZq02qHsCelO4eeV4sx_MTJJupC5qe0gDLQBtRA/edit?usp=drivesdk",
          "cachedResultName": "Sample Marketing Data"
        },
        "descriptionType": "manual",
        "toolDescription": "Extracts a sheet’s column metadata so downstream steps (querying, validation, docs) can rely on a clean, typed schema. It returns each column’s name, inferred data type, and human-readable description, plus helpful profiling hints."
      },
      "credentials": {
        "googleSheetsOAuth2Api": {
          "id": "HlBW2puZbuCCq8jJ",
          "name": "Google Sheets account 3"
        }
      },
      "typeVersion": 4.7
    },
    {
      "id": "96e7c115-ecfa-40b1-90ac-96760588a56b",
      "name": "简单记忆2",
      "type": "@n8n/n8n-nodes-langchain.memoryBufferWindow",
      "position": [
        192,
        2144
      ],
      "parameters": {},
      "typeVersion": 1.3
    },
    {
      "id": "e3fec4d3-f499-4a6e-ac00-70d2f784762c",
      "name": "OpenAI 聊天模型1",
      "type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
      "position": [
        16,
        2128
      ],
      "parameters": {
        "model": {
          "__rl": true,
          "mode": "list",
          "value": "gpt-4o",
          "cachedResultName": "gpt-4o"
        },
        "options": {}
      },
      "credentials": {
        "openAiApi": {
          "id": "4l6TDfLZVFS24g3X",
          "name": "OpenAi account 4"
        }
      },
      "typeVersion": 1.2
    },
    {
      "id": "09aab886-f1fd-447f-aded-70242aa27821",
      "name": "便签5",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -352,
        1440
      ],
      "parameters": {
        "color": 5,
        "width": 1584,
        "height": 1008,
        "content": "## 📬 需要帮助或想要自定义此工作流？"
      },
      "typeVersion": 1
    },
    {
      "id": "94911852-7bba-4178-a4bd-1461bf08d3db",
      "name": "结构化输出解析器 1",
      "type": "@n8n/n8n-nodes-langchain.outputParserStructured",
      "position": [
        528,
        2160
      ],
      "parameters": {
        "jsonSchemaExample": "{\n\t\"query\": \"sql query for google sheets\"\n}"
      },
      "typeVersion": 1.3
    },
    {
      "id": "262e5d0d-c616-4332-a0ab-f1bc9d7225ef",
      "name": "OpenAI 聊天模型2",
      "type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
      "position": [
        880,
        1840
      ],
      "parameters": {
        "model": {
          "__rl": true,
          "mode": "list",
          "value": "gpt-4.1-mini"
        },
        "options": {}
      },
      "credentials": {
        "openAiApi": {
          "id": "4l6TDfLZVFS24g3X",
          "name": "OpenAi account 4"
        }
      },
      "typeVersion": 1.2
    },
    {
      "id": "734f8a72-054b-4d04-bd9b-dc85ba160683",
      "name": "当收到聊天消息时",
      "type": "@n8n/n8n-nodes-langchain.chatTrigger",
      "position": [
        -160,
        1744
      ],
      "webhookId": "edca0f0a-77c3-43e5-8ece-e514a29446f5",
      "parameters": {
        "options": {}
      },
      "typeVersion": 1.3
    },
    {
      "id": "be5f8fc5-0d06-4acf-8fa8-efca6310f0ec",
      "name": "从 Google Sheets 获取数据",
      "type": "n8n-nodes-base.httpRequest",
      "position": [
        640,
        1888
      ],
      "parameters": {
        "url": "={{ $json.output.query }}",
        "options": {},
        "sendQuery": true,
        "queryParameters": {
          "parameters": [
            {}
          ]
        }
      },
      "typeVersion": 4.2
    },
    {
      "id": "71029e6d-3933-49a0-b31b-1461ef3a176f",
      "name": "写入表格输出1",
      "type": "@n8n/n8n-nodes-langchain.agent",
      "position": [
        864,
        1616
      ],
      "parameters": {
        "text": "={{ $json.data }}",
        "options": {
          "systemMessage": "write this into one table. output as dimensions, then metrics"
        },
        "promptType": "define"
      },
      "typeVersion": 2.2
    },
    {
      "id": "43d7c1a8-0fc0-4877-ab9c-908c3836239a",
      "name": "便签 12",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -1920,
        1440
      ],
      "parameters": {
        "width": 752,
        "height": 1008,
        "content": "## 分步设置说明"
      },
      "typeVersion": 1
    },
    {
      "id": "0081f73b-ae42-424d-8cd1-9f9e0b30c64f",
      "name": "便签13",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -1136,
        1440
      ],
      "parameters": {
        "width": 752,
        "height": 1008,
        "content": ""
      },
      "typeVersion": 1
    }
  ],
  "pinData": {},
  "connections": {
    "AI Agent3": {
      "main": [
        [
          {
            "node": "Get Data from Google Sheet",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Simple Memory2": {
      "ai_memory": [
        [
          {
            "node": "AI Agent3",
            "type": "ai_memory",
            "index": 0
          }
        ]
      ]
    },
    "Get Column Info2": {
      "ai_tool": [
        [
          {
            "node": "AI Agent3",
            "type": "ai_tool",
            "index": 0
          }
        ]
      ]
    },
    "OpenAI Chat Model1": {
      "ai_languageModel": [
        [
          {
            "node": "AI Agent3",
            "type": "ai_languageModel",
            "index": 0
          }
        ]
      ]
    },
    "OpenAI Chat Model2": {
      "ai_languageModel": [
        [
          {
            "node": "Write into Table Output1",
            "type": "ai_languageModel",
            "index": 0
          }
        ]
      ]
    },
    "Structured Output Parser1": {
      "ai_outputParser": [
        [
          {
            "node": "AI Agent3",
            "type": "ai_outputParser",
            "index": 0
          }
        ]
      ]
    },
    "Get Data from Google Sheet": {
      "main": [
        [
          {
            "node": "Write into Table Output1",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "When chat message received": {
      "main": [
        [
          {
            "node": "AI Agent3",
            "type": "main",
            "index": 0
          }
        ]
      ]
    }
  }
}