Natural Language Query to Structured Data Retrieval with LLM

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 →

{
  "name": "Natural Language Query to Structured Data Retrieval with LLM",
  "nodes": [
    {
      "parameters": {
        "formTitle": "Ask a Question About Your Dataset",
        "formDescription": "Enter your question in natural language (e.g., \"Which students scored more than 80 marks in NLP?\")",
        "formFields": {
          "values": [
            {
              "fieldLabel": "Your Question",
              "fieldType": "textarea",
              "fieldName": "question",
              "placeholder": "e.g., Which students scored more than 80 marks in NLP?",
              "requiredField": true
            },
            {
              "fieldLabel": "Upload Dataset (Excel/CSV)",
              "fieldType": "file",
              "fieldName": "dataset",
              "multipleFiles": false,
              "acceptFileTypes": ".xlsx,.xls,.csv",
              "requiredField": true
            }
          ]
        },
        "responseMode": "lastNode",
        "options": {}
      },
      "id": "0e464341-ff06-4c37-9327-afa17cd696b7",
      "name": "Question Form",
      "type": "n8n-nodes-base.formTrigger",
      "typeVersion": 2.5,
      "position": [
        240,
        304
      ]
    },
    {
      "parameters": {
        "operation": "xlsx",
        "binaryPropertyName": "dataset",
        "options": {
          "headerRow": true,
          "includeEmptyCells": false
        }
      },
      "id": "58af128f-5cdd-40c3-8635-053edf9d8305",
      "name": "Load Dataset",
      "type": "n8n-nodes-base.extractFromFile",
      "typeVersion": 1.1,
      "position": [
        464,
        304
      ]
    },
    {
      "parameters": {
        "assignments": {
          "assignments": [
            {
              "id": "1",
              "name": "response",
              "value": "={{ $json.content.parts[0].text }}",
              "type": "string"
            }
          ]
        },
        "options": {}
      },
      "id": "04d2e82c-828a-48ef-8cd3-73073a20f9be",
      "name": "Format Response",
      "type": "n8n-nodes-base.set",
      "typeVersion": 3.4,
      "position": [
        1248,
        304
      ]
    },
    {
      "parameters": {
        "modelId": {
          "__rl": true,
          "value": "models/gemini-2.5-flash",
          "mode": "list",
          "cachedResultName": "models/gemini-2.5-flash"
        },
        "messages": {
          "values": [
            {
              "content": "=User Question: {{ $(\"Question Form\").item.json.question }}\n\nComplete Dataset:\n{{ JSON.stringify($(\"Load Dataset\").all().slice(0, 50).map(i => i.json), null, 2) }}\n\nPlease analyze the dataset and answer the user's question. Provide a clear, conversational answer with the matching records."
            }
          ]
        },
        "builtInTools": {},
        "options": {
          "systemMessage": "You are a helpful data analyst assistant. Analyze the provided dataset and answer the user's natural language question. Filter and present the relevant records in a clear, conversational format.",
          "maxOutputTokens": 2048,
          "temperature": 0.3
        }
      },
      "id": "558e71fd-a098-4109-b4ff-f67217b45b10",
      "name": "Query Dataset with Gemini",
      "type": "@n8n/n8n-nodes-langchain.googleGemini",
      "typeVersion": 1.1,
      "position": [
        896,
        304
      ],
      "credentials": {
        "googlePalmApi": {
          "name": "<your credential>"
        }
      }
    },
    {
      "parameters": {
        "jsCode": "// Loop over input items and add a new field called 'myNewField' to the JSON of each one\n\nreturn [$input.all()[0]];\n"
      },
      "type": "n8n-nodes-base.code",
      "typeVersion": 2,
      "position": [
        672,
        304
      ],
      "id": "8a32b47a-4894-4e67-98e5-219b29e05907",
      "name": "Code in JavaScript"
    }
  ],
  "connections": {
    "Question Form": {
      "main": [
        [
          {
            "node": "Load Dataset",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Load Dataset": {
      "main": [
        [
          {
            "node": "Code in JavaScript",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Query Dataset with Gemini": {
      "main": [
        [
          {
            "node": "Format Response",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Code in JavaScript": {
      "main": [
        [
          {
            "node": "Query Dataset with Gemini",
            "type": "main",
            "index": 0
          }
        ]
      ]
    }
  },
  "active": true,
  "settings": {
    "executionOrder": "v1",
    "binaryMode": "separate",
    "availableInMCP": false
  },
  "versionId": "5c3d4a51-3b56-4e5b-8225-4d9ed3132ebb",
  "meta": {
    "templateCredsSetupCompleted": true
  },
  "id": "aLvRITXk6ZGWq7PC",
  "tags": []
}

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.

• googlePalmApi

About this workflow

Natural Language Query to Structured Data Retrieval with LLM. Uses formTrigger, googleGemini. Event-driven trigger; 5 nodes.

Source: https://github.com/abhitrueprogrammer/nlp-tasks/blob/c77c7b64bc7bbe1519d6e06fdf262cd905f9ffc0/n8n/n8n-complex.json — 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.