AI Message Classifier for TableFlow

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": "TableFlow \u2014 AI Message Classifier",
  "nodes": [
    {
      "parameters": {
        "httpMethod": "POST",
        "path": "ai-classifier",
        "responseMode": "responseNode",
        "options": {}
      },
      "id": "n8n-node-webhook",
      "name": "Classifier Webhook",
      "type": "n8n-nodes-base.webhook",
      "typeVersion": 1,
      "position": [
        240,
        300
      ]
    },
    {
      "parameters": {
        "jsCode": "const body = $input.first().json.body;\nconst message = body.message || '';\nconst today = new Date().toISOString().slice(0, 10);\n\nconst systemPrompt = `You are a restaurant reservation assistant. Extract structured information from incoming messages.\n\nToday's date is ${today}.\n\nAlways respond with valid JSON only \u2014 no prose, no markdown, no code fences.\n\nRequired JSON schema:\n{\n  \"intent\": \"new_reservation\" | \"cancellation\" | \"modification\" | \"inquiry\" | \"other\",\n  \"name\": string | null,\n  \"phone\": string | null,\n  \"email\": string | null,\n  \"party_size\": number | null,\n  \"datetime\": string | null,\n  \"notes\": string | null,\n  \"confidence\": number\n}\n\nRules:\n- intent: classify the message purpose\n- datetime: ISO-8601 format (YYYY-MM-DDTHH:mm:00.000Z), infer from context if relative (\"tomorrow evening\" = next day at 20:00)\n- party_size: integer, extract from \"for 4\", \"table for two\", \"party of 5\", etc.\n- confidence: 0.0\u20131.0, how confident you are in the extraction\n- Use null for fields that cannot be determined from the message`;\n\nreturn [{\n  json: {\n    system_prompt: systemPrompt,\n    user_message: message,\n    channel: body.channel || 'unknown',\n    from: body.from || ''\n  }\n}];"
      },
      "id": "n8n-node-build-prompt",
      "name": "Build Prompt",
      "type": "n8n-nodes-base.code",
      "typeVersion": 2,
      "position": [
        460,
        300
      ]
    },
    {
      "parameters": {
        "url": "https://api.openai.com/v1/chat/completions",
        "authentication": "genericCredentialType",
        "genericAuthType": "httpHeaderAuth",
        "sendHeaders": true,
        "headerParameters": {
          "parameters": [
            {
              "name": "Authorization",
              "value": "=Bearer {{ $env.OPENAI_API_KEY }}"
            },
            {
              "name": "Content-Type",
              "value": "application/json"
            }
          ]
        },
        "sendBody": true,
        "specifyBody": "json",
        "jsonBody": "={\n  \"model\": \"gpt-4o-mini\",\n  \"temperature\": 0,\n  \"max_tokens\": 300,\n  \"response_format\": { \"type\": \"json_object\" },\n  \"messages\": [\n    { \"role\": \"system\", \"content\": {{ JSON.stringify($json.system_prompt) }} },\n    { \"role\": \"user\",   \"content\": {{ JSON.stringify($json.user_message) }} }\n  ]\n}",
        "options": {
          "timeout": 20000
        }
      },
      "id": "n8n-node-openai",
      "name": "Call OpenAI",
      "type": "n8n-nodes-base.httpRequest",
      "typeVersion": 4,
      "position": [
        680,
        300
      ],
      "credentials": {
        "httpHeaderAuth": {
          "name": "<your credential>"
        }
      }
    },
    {
      "parameters": {
        "jsCode": "const response = $input.first().json;\n\nlet parsed;\ntry {\n  const content = response.choices?.[0]?.message?.content || '{}';\n  parsed = JSON.parse(content);\n} catch (err) {\n  parsed = {\n    intent: 'other',\n    name: null,\n    phone: null,\n    email: null,\n    party_size: null,\n    datetime: null,\n    notes: null,\n    confidence: 0\n  };\n}\n\n// Ensure confidence is a valid number\nparsed.confidence = Math.min(1, Math.max(0, Number(parsed.confidence) || 0));\n\n// Validate datetime format\nif (parsed.datetime) {\n  const d = new Date(parsed.datetime);\n  if (isNaN(d.getTime())) parsed.datetime = null;\n}\n\nreturn [{ json: parsed }];"
      },
      "id": "n8n-node-parse",
      "name": "Parse AI Response",
      "type": "n8n-nodes-base.code",
      "typeVersion": 2,
      "position": [
        900,
        300
      ]
    },
    {
      "parameters": {
        "respondWith": "json",
        "responseBody": "={{ JSON.stringify($json) }}"
      },
      "id": "n8n-node-respond",
      "name": "Return Classification",
      "type": "n8n-nodes-base.respondToWebhook",
      "typeVersion": 1,
      "position": [
        1120,
        300
      ]
    }
  ],
  "connections": {
    "Classifier Webhook": {
      "main": [
        [
          {
            "node": "Build Prompt",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Build Prompt": {
      "main": [
        [
          {
            "node": "Call OpenAI",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Call OpenAI": {
      "main": [
        [
          {
            "node": "Parse AI Response",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Parse AI Response": {
      "main": [
        [
          {
            "node": "Return Classification",
            "type": "main",
            "index": 0
          }
        ]
      ]
    }
  },
  "settings": {
    "executionOrder": "v1"
  },
  "meta": {
    "notes": "Accepts a free-text message and returns structured JSON with intent, name, phone, email, party_size, datetime, and confidence. Uses gpt-4o-mini with json_object response format. Webhook URL: POST http://<n8n-host>:5678/webhook/ai-classifier"
  }
}

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.

• httpHeaderAuth

About this workflow

TableFlow — AI Message Classifier. Uses httpRequest. Webhook trigger; 5 nodes.

Source: https://github.com/SanchezAlbaladejoPablo/tableflow/blob/72a4c27d1469a2f43764787cd24332c050448e9d/automations/n8n/ai-classifier.json — original creator credit. Request a take-down →

More Web Scraping workflows → · Browse all categories →

Related workflows

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