Robert Breen This workflow corresponds to n8n.io template #7154 — we link there as the canonical source.
This workflow follows the Agent → OpenAI Chat recipe pattern — see all workflows that pair these two integrations.
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 →
{
"meta": {
"templateCredsSetupCompleted": true
},
"nodes": [
{
"id": "bd247a8c-fa46-4b86-ad4b-977b73c35d4e",
"name": "OpenAI Chat Model1",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
-540,
1340
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "gpt-4o-mini"
},
"options": {}
},
"credentials": {
"openAiApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.2
},
{
"id": "e97ac821-6da8-4bea-875c-a4ccdf0d53b2",
"name": "Tool: Inject Creativity1",
"type": "@n8n/n8n-nodes-langchain.toolThink",
"position": [
-440,
1520
],
"parameters": {},
"typeVersion": 1
},
{
"id": "312c5a50-8c9b-4aa9-b3b4-9b682e757bc1",
"name": "Structured Output Parser",
"type": "@n8n/n8n-nodes-langchain.outputParserStructured",
"position": [
-300,
1320
],
"parameters": {
"jsonSchemaExample": "{\n \"column1\": [\n \"2025-08-01\",\n \"2025-08-02\",\n \"2025-08-03\"\n ],\n \"column2\": [\n \"Instagram\",\n \"LinkedIn\",\n \"Twitter\"\n ],\n \"column3\": [\n \"Image Post\",\n \"Blog Link\",\n \"Video Snippet\"\n ],\n \"column4\": [\n \"Workflow Automation\",\n \"AI Agent Demo\",\n \"Case Study\"\n ],\n \"column5\": [\n \"Alice\",\n \"Bob\",\n \"Charlie\"\n ]\n}\n"
},
"typeVersion": 1.2
},
{
"id": "975b4e46-656d-4128-88c9-c8dc385e2fb8",
"name": "OpenAI Chat Model2",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
240,
1740
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "gpt-4o-mini"
},
"options": {}
},
"credentials": {
"openAiApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.2
},
{
"id": "052b20f4-8724-41fe-b2c1-937ece4003ea",
"name": "Structured Output Parser1",
"type": "@n8n/n8n-nodes-langchain.outputParserStructured",
"position": [
420,
1760
],
"parameters": {
"jsonSchemaExample": "{\n \"columnnames\": [\n \"first\",\n \"second\",\n \"third\"\n ]\n}\n"
},
"typeVersion": 1.2
},
{
"id": "a523d13f-653d-4a91-a672-b0bd5d558f27",
"name": "Run Workflow",
"type": "n8n-nodes-base.manualTrigger",
"position": [
-1080,
1440
],
"parameters": {},
"typeVersion": 1
},
{
"id": "ae8bc30f-c16e-435c-a875-8c02287d4855",
"name": "Sticky Note4",
"type": "n8n-nodes-base.stickyNote",
"position": [
-1160,
540
],
"parameters": {
"color": 5,
"width": 420,
"height": 1340,
"content": "### \ud83e\udd47 Step 1: Setup OpenAI API Credentials\n\n1. Go to [https://platform.openai.com/account/api-keys](https://platform.openai.com/account/api-keys)\n2. Click **\u201cCreate new secret key\u201d**\n3. Copy your API key\n4. In n8n:\n - Go to **Credentials**\n - Click **\u201cNew Credential\u201d**\n - Select `OpenAI API`\n - Paste your API key\n - Name it something like `OpenAI account`\n\n\u27a1\ufe0f You will use this credential in:\n- `OpenAI Chat Model1`\n- `OpenAI Chat Model2`\n\n---\n\n### \ud83e\udd48 Step 2: Add a Manual Trigger Node\n\n- Type: `Manual Trigger`\n- Purpose: Starts the workflow manually for testing\n- No configuration required\n\n---\n\n### \ud83e\udd49 Step 3: Set Your Topic (Set Node)\n\n- Node: `Set Topic to Search`\n- Type: `Set`\n- Add a new string field:\n - Name: `Topic`\n - Value: e.g., `n8n use cases`\n\nThis is the topic the workflow will generate data for.\n\n---"
},
"typeVersion": 1
},
{
"id": "42d4c54f-bb88-4c0d-a31c-0beb1814b0b7",
"name": "Sticky Note5",
"type": "n8n-nodes-base.stickyNote",
"position": [
-700,
540
],
"parameters": {
"color": 6,
"width": 780,
"height": 1340,
"content": "\n### \u2728 Step 4: Generate Structured Data\n- **LangChain Agent** node `Generate Random Data`\n- Connect to **OpenAI Chat Model1** and **Tool: Inject Creativity1** \n- System prompt: instruct AI to output 5 columns of realistic values in JSON \n\n### \ud83d\udd27 Step 5: Parse AI Output\n- **Structured Output Parser** to validate JSON \n\n### \ud83d\udd04 Step 6: Flatten Data\n- **Code** node `Outpt all Data to One Field` \n- Joins all values into a comma-separated string for column naming"
},
"typeVersion": 1
},
{
"id": "4c48281e-5c67-4b71-8b20-23ff69d50582",
"name": "Sticky Note7",
"type": "n8n-nodes-base.stickyNote",
"position": [
100,
540
],
"parameters": {
"color": 3,
"width": 640,
"height": 1340,
"content": "### \ud83e\udde0 Step 7: Generate Column Names\n- **LangChain Agent** `Generate Column Names` \n- Connect to **OpenAI Chat Model2** \n- Prompt: infer 5 column names from the string \n\n### \ud83d\udd22 Step 8: Pivot Names Row\n- **Code** node `Pivot Column Names` transforms array into `{ column1: name1, \u2026 }`\n"
},
"typeVersion": 1
},
{
"id": "6c6774de-9bb1-4c19-b121-fc6ccb2dcdc5",
"name": "Sticky Note8",
"type": "n8n-nodes-base.stickyNote",
"position": [
760,
540
],
"parameters": {
"color": 2,
"width": 1460,
"height": 1340,
"content": "### \ud83e\ude93 Step 9: Split Columns\n- 5 `SplitOut` nodes to break each array back into rows per column\n\n### \ud83d\udd17 Step 10: Merge Rows\n- **Merge** node `Merge Columns together` using `combineByPosition` \n\n### \ud83c\udff7\ufe0f Step 11: Rename Columns\n- **Set** node `Rename Columns` assigns the AI-generated names to each column\n\n### \ud83d\udd17 Step 12: Final Output\n- **Merge** `Append Column Names` combines data and header row"
},
"typeVersion": 1
},
{
"id": "014107b1-ff36-438f-8c88-8196ad4f48f7",
"name": "Sticky Note10",
"type": "n8n-nodes-base.stickyNote",
"position": [
-1160,
340
],
"parameters": {
"width": 3380,
"content": "## \ud83d\udcec Need Help or Want to Customize This?\n\ud83d\udce7 [robert@ynteractive.com](mailto:robert@ynteractive.com) \n\ud83d\udd17 [LinkedIn](https://www.linkedin.com/in/robert-breen-29429625/)"
},
"typeVersion": 1
},
{
"id": "cc11756c-f453-4cce-b148-49879bced88b",
"name": "Set Topic to Search",
"type": "n8n-nodes-base.set",
"position": [
-920,
1620
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "23fed7d1-74bb-487a-bd39-59abb02b9373",
"name": "Topic",
"type": "string",
"value": "n8n use cases"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "7b0ab29e-8d4e-4576-8933-101142364284",
"name": "Generate Random Data",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
-520,
1020
],
"parameters": {
"text": "=idea: {{ $json.Topic }}",
"options": {
"systemMessage": "You are a tool that generates structured sample data in JSON format.\n\nWhen given a topic or a description of the type of data the user needs (e.g., \"marketing campaigns\", \"customer feedback\", \"blog ideas\"), do the following:\n\n1. Identify 5 relevant columns that would make up a realistic dataset for the topic.\n2. Generate 3\u20135 realistic values for each column.\n3. Output the result in a JSON object using the following structure:\n - Each key should be labeled as \"columnX (Column Name)\" where X is the column number from 1 to 5.\n - Each value should be a list of 3\u20135 strings representing data for that column.\n\nDo not explain your output. Do not include anything outside the JSON.\n\nOutput 5 columns of data like this. \n\nExample format:\n{\n \"column1 (Date)\": [\n \"2025-08-01\",\n \"2025-08-02\",\n \"2025-08-03\"\n ],\n \"column2 (Platform)\": [\n \"Instagram\",\n \"LinkedIn\",\n \"Twitter\"\n ],\n \"column3 (Content Type)\": [\n \"Image Post\",\n \"Blog Link\",\n \"Video Snippet\"\n ],\n \"column4 (Topic)\": [\n \"Workflow Automation\",\n \"AI Agent Demo\",\n \"Case Study\"\n ],\n \"column5 (Owner)\": [\n \"Alice\",\n \"Bob\",\n \"Charlie\"\n ]\n}\n\n\nMake sure the data is contextually relevant to the user's input.\n"
},
"promptType": "define",
"hasOutputParser": true
},
"typeVersion": 2
},
{
"id": "843589b2-36cd-4c3e-b710-c683de9ebdd8",
"name": "Outpt all Data to One Field",
"type": "n8n-nodes-base.code",
"position": [
-120,
1260
],
"parameters": {
"jsCode": "// Get the object from input\nconst data = $input.first().json.output;\n\n// Flatten all column values into one array\nconst allValues = Object.values(data).flat();\n\n// Join all values with commas\nconst result = allValues.join(', ');\n\n// Return the final text as a single field\nreturn [\n {\n json: {\n text: result\n }\n }\n];\n"
},
"typeVersion": 2
},
{
"id": "87405d02-b882-403e-bf5a-9d125b1536b8",
"name": "Generate Column Names",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
260,
1480
],
"parameters": {
"text": "=output: {{ $json.text }}",
"options": {
"systemMessage": "Take the input and output relevent column names for the data. there are 5 columns. give each of them a name that makes sense for the values in the column. "
},
"promptType": "define",
"hasOutputParser": true
},
"typeVersion": 2
},
{
"id": "419ce3c1-1fe5-4d8a-8330-e8afca8c94f5",
"name": "Pivot Column Names",
"type": "n8n-nodes-base.code",
"position": [
600,
1660
],
"parameters": {
"jsCode": "const columnNames = $input.first().json.output.columnnames;\n\n// Build a single row with column1, column2, etc. as keys and names as values\nconst row = {};\n\ncolumnNames.forEach((name, index) => {\n row[`column${index + 1}`] = name;\n});\n\nreturn [\n { json: row }\n];\n"
},
"typeVersion": 2
},
{
"id": "1b3f150e-60a5-466b-9874-c2f98f411eaa",
"name": "Split Column 1",
"type": "n8n-nodes-base.splitOut",
"position": [
1100,
1020
],
"parameters": {
"options": {},
"fieldToSplitOut": "output.column1"
},
"typeVersion": 1
},
{
"id": "dfee12fe-f157-48e0-8b6c-71a1bc0eeac4",
"name": "Split Column 2",
"type": "n8n-nodes-base.splitOut",
"position": [
920,
1120
],
"parameters": {
"options": {},
"fieldToSplitOut": "output.column2"
},
"typeVersion": 1
},
{
"id": "effd383c-a799-4500-be47-1052549e8cf9",
"name": "Split Column 3",
"type": "n8n-nodes-base.splitOut",
"position": [
1120,
1240
],
"parameters": {
"options": {},
"fieldToSplitOut": "output.column3"
},
"typeVersion": 1
},
{
"id": "e346941b-8b34-467c-ae42-02dd0c6f45e8",
"name": "Split Column 4",
"type": "n8n-nodes-base.splitOut",
"position": [
880,
1320
],
"parameters": {
"options": {},
"fieldToSplitOut": "output.column4"
},
"typeVersion": 1
},
{
"id": "6a7384d7-f615-4ed0-8f20-8a6a010f4984",
"name": "Split Column 5",
"type": "n8n-nodes-base.splitOut",
"position": [
1100,
1480
],
"parameters": {
"options": {},
"fieldToSplitOut": "output.column5"
},
"typeVersion": 1
},
{
"id": "daeede53-e473-47bd-836e-9bbabba95697",
"name": "Merge Columns together",
"type": "n8n-nodes-base.merge",
"position": [
1380,
1260
],
"parameters": {
"mode": "combine",
"options": {},
"combineBy": "combineByPosition",
"numberInputs": 5
},
"typeVersion": 3.2
},
{
"id": "b703dffd-ee3e-41da-b883-34e806fb56d9",
"name": "Rename Columns",
"type": "n8n-nodes-base.set",
"position": [
1620,
1300
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "3b6cd7c0-b2ab-48bd-9d3d-c6f577d43a32",
"name": "column1",
"type": "string",
"value": "={{ $('Split Column 1').item.json['output.column1'] }}"
},
{
"id": "e19027d6-5ebd-43ed-922c-bb5183844875",
"name": "column2",
"type": "string",
"value": "={{ $('Split Column 2').item.json['output.column2'] }}"
},
{
"id": "81339019-9a39-4e7c-a3a1-53e7370ce7c1",
"name": "column3",
"type": "string",
"value": "={{ $('Split Column 3').item.json['output.column3'] }}"
},
{
"id": "7cfb8fa4-e25c-49e6-96dc-66da82f95882",
"name": "column4",
"type": "string",
"value": "={{ $('Split Column 4').item.json['output.column4'] }}"
},
{
"id": "3301a0dc-ff0c-42a1-8df0-e3dcafed4001",
"name": "column5",
"type": "string",
"value": "={{ $('Split Column 5').item.json['output.column5'] }}"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "587fbb98-a687-43e6-89dc-dd71d98b5211",
"name": "Append Column Names",
"type": "n8n-nodes-base.merge",
"position": [
1840,
1680
],
"parameters": {},
"typeVersion": 3.2
}
],
"connections": {
"Run Workflow": {
"main": [
[
{
"node": "Set Topic to Search",
"type": "main",
"index": 0
}
]
]
},
"Rename Columns": {
"main": [
[
{
"node": "Append Column Names",
"type": "main",
"index": 1
}
]
]
},
"Split Column 1": {
"main": [
[
{
"node": "Merge Columns together",
"type": "main",
"index": 0
}
]
]
},
"Split Column 2": {
"main": [
[
{
"node": "Merge Columns together",
"type": "main",
"index": 1
}
]
]
},
"Split Column 3": {
"main": [
[
{
"node": "Merge Columns together",
"type": "main",
"index": 2
}
]
]
},
"Split Column 4": {
"main": [
[
{
"node": "Merge Columns together",
"type": "main",
"index": 3
}
]
]
},
"Split Column 5": {
"main": [
[
{
"node": "Merge Columns together",
"type": "main",
"index": 4
}
]
]
},
"OpenAI Chat Model1": {
"ai_languageModel": [
[
{
"node": "Generate Random Data",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"OpenAI Chat Model2": {
"ai_languageModel": [
[
{
"node": "Generate Column Names",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Pivot Column Names": {
"main": [
[
{
"node": "Append Column Names",
"type": "main",
"index": 0
}
]
]
},
"Append Column Names": {
"main": [
[]
]
},
"Set Topic to Search": {
"main": [
[
{
"node": "Generate Random Data",
"type": "main",
"index": 0
}
]
]
},
"Generate Random Data": {
"main": [
[
{
"node": "Split Column 1",
"type": "main",
"index": 0
},
{
"node": "Split Column 2",
"type": "main",
"index": 0
},
{
"node": "Split Column 3",
"type": "main",
"index": 0
},
{
"node": "Split Column 4",
"type": "main",
"index": 0
},
{
"node": "Split Column 5",
"type": "main",
"index": 0
},
{
"node": "Outpt all Data to One Field",
"type": "main",
"index": 0
}
]
]
},
"Generate Column Names": {
"main": [
[
{
"node": "Pivot Column Names",
"type": "main",
"index": 0
}
]
]
},
"Merge Columns together": {
"main": [
[
{
"node": "Rename Columns",
"type": "main",
"index": 0
}
]
]
},
"Structured Output Parser": {
"ai_outputParser": [
[
{
"node": "Generate Random Data",
"type": "ai_outputParser",
"index": 0
}
]
]
},
"Tool: Inject Creativity1": {
"ai_tool": [
[
{
"node": "Generate Random Data",
"type": "ai_tool",
"index": 0
}
]
]
},
"Structured Output Parser1": {
"ai_outputParser": [
[
{
"node": "Generate Column Names",
"type": "ai_outputParser",
"index": 0
}
]
]
},
"Outpt all Data to One Field": {
"main": [
[
{
"node": "Generate Column Names",
"type": "main",
"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.
openAiApi
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
This n8n workflow dynamically generates a realistic sample dataset based on a single topic you provide. It uses OpenAI (via LangChain) and n8n’s built-in nodes to: Generate structured JSON data for 5 columns with 3–5 values each Flatten that data into a single text blob Infer…
Source: https://n8n.io/workflows/7154/ — original creator credit. Request a take-down →
Related workflows
Workflows that share integrations, category, or trigger type with this one. All free to copy and import.
🎯 Create viral TikToks, Shorts, Reels, podcasts, and ASMR videos in minutes — all on autopilot.
OpenAI, HTTP Request, Form Trigger +7
Generate AI viral videos with NanoBanana & VEO3, shared on socials via Blotato 2. Uses @blotato/n8n-nodes-blotato, googleSheets, lmChatOpenAi, toolThink. Event-driven trigger; 94 nodes.
@Blotato/N8N Nodes Blotato, Google Sheets, OpenAI Chat +9
This template is designed for marketers, content creators, and e-commerce brands who want to automate the creation of professional ad videos at scale. It’s ideal for teams looking to generate consiste
Telegram, Telegram Trigger, Google Drive +8
💥 Automate video ads with NanoBanana, Seedream 4, ChatGPT Image and Veo 3 - VIDE. Uses telegram, telegramTrigger, googleDrive, agent. Event-driven trigger; 63 nodes.
Telegram, Telegram Trigger, Google Drive +8
💥 Automate video ads with NanoBanana, Seedream 4, ChatGPT Image and Veo 3 - VIDE. Uses telegram, telegramTrigger, googleDrive, agent. Event-driven trigger; 63 nodes.
Telegram, Telegram Trigger, Google Drive +8