JKingma This workflow corresponds to n8n.io template #8475 — we link there as the canonical source.
This workflow follows the Chainllm → HTTP Request 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 →
{
"id": "LVoDtVFQ069sOYcC",
"meta": {
"templateCredsSetupCompleted": true
},
"name": "Automated Product Description Generation for Adobe Commerce (Magento 2)",
"tags": [],
"nodes": [
{
"id": "7272f57f-3ebd-4a5d-a3ae-1c6a16e0a140",
"name": "When clicking \u2018Test workflow\u2019",
"type": "n8n-nodes-base.manualTrigger",
"position": [
-1504,
16
],
"parameters": {},
"typeVersion": 1
},
{
"id": "1e10c1ee-3777-466e-ab0f-a211046f4114",
"name": "Save configurable",
"type": "n8n-nodes-base.httpRequest",
"position": [
64,
16
],
"parameters": {
"url": "=https://www.example.com/rest/V1/products/{{ encodeURIComponent($('get Product without description').first().json.items[0].sku) }}",
"method": "PUT",
"options": {},
"jsonBody": "={\n \"product\": {\n \"custom_attributes\": [{\n \"attribute_code\": \"description\",\n \"value\": \"{{ $json.description.replaceAll('\\n',' ').replaceAll('\"\"','\\\\\"') }}\"\n }]\n }\n}",
"sendBody": true,
"specifyBody": "json",
"authentication": "predefinedCredentialType",
"nodeCredentialType": "magento2Api"
},
"credentials": {
"magento2Api": {
"name": "<your credential>"
}
},
"typeVersion": 4.2
},
{
"id": "5373fa9b-e857-435f-a3ae-08978e7717df",
"name": "set Description",
"type": "n8n-nodes-base.set",
"position": [
-128,
16
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "c21f389d-4104-4a6f-b616-324abb221798",
"name": "description",
"type": "string",
"value": "={{ $json.text }}"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "e127791d-b564-4c23-86a2-a55cbe1bad49",
"name": "Basic LLM Chain",
"type": "@n8n/n8n-nodes-langchain.chainLlm",
"position": [
-528,
16
],
"parameters": {
"text": "=Write an English product description based on this Magento2 attributes. Please use html paragraphs and breaks only. Ensure the text is grammatically correct, natural, and tailored to the ... market. Use terminology that is common in this sector, and avoid overly literal or informal style. \n\nWrite in a consistent style suitable for webshop product descriptions: professional, clear, and customer-oriented, while keeping a persuasive tone aimed at ...\n\nAdditionally, make the text SEO-friendly : include relevant keywords naturally without unnecessary repetition, and ensure the text reads smoothly while supporting search visibility.\n\nHere is the Magento product: {{ JSON.stringify($json) }}",
"batching": {},
"promptType": "define"
},
"typeVersion": 1.7
},
{
"id": "487315aa-572d-40a4-ba61-a091b8157061",
"name": "Azure OpenAI Chat Model",
"type": "@n8n/n8n-nodes-langchain.lmChatAzureOpenAi",
"position": [
-528,
192
],
"parameters": {
"model": "model-router",
"options": {}
},
"credentials": {
"azureOpenAiApi": {
"name": "<your credential>"
}
},
"typeVersion": 1
},
{
"id": "5198d2bb-89db-401a-bf35-21884378c2f5",
"name": "Wait until nodes ready",
"type": "n8n-nodes-base.merge",
"position": [
-1040,
16
],
"parameters": {},
"typeVersion": 3.2
},
{
"id": "3807a3cf-d7aa-4315-8d25-cc6386b1181a",
"name": "get Product without description",
"type": "n8n-nodes-base.httpRequest",
"position": [
-1264,
-128
],
"parameters": {
"url": "=https://www.example.com/rest/V1/products",
"options": {},
"sendQuery": true,
"authentication": "predefinedCredentialType",
"queryParameters": {
"parameters": [
{
"name": "searchCriteria[filter_groups][0][filters][0][field]",
"value": "description"
},
{
"name": "searchCriteria[filter_groups][0][filters][0][condition_type]",
"value": "null"
},
{
"name": "searchCriteria[pageSize]",
"value": "1"
}
]
},
"nodeCredentialType": "magento2Api"
},
"credentials": {
"magento2Api": {
"name": "<your credential>"
}
},
"typeVersion": 4.2
},
{
"id": "9c551919-8e15-44bb-937b-3ae06dc5ffc2",
"name": "Prepare attributes",
"type": "n8n-nodes-base.code",
"position": [
-720,
16
],
"parameters": {
"jsCode": "// Assuming the JSON data is stored in a variable called \"jsonData\"\nconst product = $(\"get Product without description\").first().json.items[0];\nconst data = $('Get attributes and options').first().json.items; // Extract the items containing attributes and options\nconst customAttributes = product.custom_attributes; // Extract the custom attributes\n//console.log(customAttributes);\n// Create a map of attribute_code to their options (for select-type attributes)\nconst attributeOptionsMap = {};\nfor (const attribute of data) {\n if (attribute.options) {\n attributeOptionsMap[attribute.attribute_code] = attribute.options;\n }\n}\n// Loop over all custom attributes to check if they exist in the attributeOptionsMap\nfor (const customAttribute of customAttributes) {\n const attributeCode = customAttribute.attribute_code;\n //console.log(attributeCode);\n\n // Check if this custom attribute exists in the attributeOptionsMap (i.e., it has options)\n if (attributeOptionsMap[attributeCode]) {\n \n const attributeOptions = attributeOptionsMap[attributeCode];\n const value = customAttribute.value;\n //console.log(value);\n\n // Find the label corresponding to the current value\n const matchingOption = attributeOptions.find(option => option.value === value);\n if (matchingOption) {\n // Replace the value with the label\n customAttribute.value = matchingOption.label;\n }\n }\n}\n// After this, customAttributes will have all applicable values replaced by their labels\nproduct.custom_attributes = customAttributes;\nreturn product;"
},
"typeVersion": 2
},
{
"id": "88d7173a-e816-4530-87e5-6be43f44b5b2",
"name": "Get attributes and options",
"type": "n8n-nodes-base.httpRequest",
"position": [
-1264,
144
],
"parameters": {
"url": "https://www.example.com/rest/V1/products/attributes?searchCriteria[filter_groups][0][filters][0][field]=frontend_input&searchCriteria[filter_groups][0][filters][0][value]=select&fields=items[frontend_input,attribute_code,default_frontend_label,options]",
"options": {},
"authentication": "predefinedCredentialType",
"nodeCredentialType": "magento2Api"
},
"credentials": {
"magento2Api": {
"name": "<your credential>"
}
},
"executeOnce": true,
"typeVersion": 4.2
},
{
"id": "50edd904-37c6-44cb-860c-023041f51892",
"name": "Sticky Note",
"type": "n8n-nodes-base.stickyNote",
"position": [
-1552,
-336
],
"parameters": {
"color": 4,
"width": 720,
"height": 704,
"content": "### Step 1: Get Product Information from Magento\n- Retrieves a single product that is missing a description. \n- Reads all product attributes through the Magento API. "
},
"typeVersion": 1
},
{
"id": "7f7d74f8-2203-4268-a342-193b2c01eac9",
"name": "Sticky Note1",
"type": "n8n-nodes-base.stickyNote",
"position": [
-800,
-336
],
"parameters": {
"color": 6,
"width": 560,
"height": 704,
"content": "### Step 2: Generate Description with LLM\n- Resolves attribute options into **human-readable labels** (e.g. size = `3` \u2192 \"Large\"). \n- Prepares structured product data for use in the LLM prompt. \n- Passes the **readable product attributes** to an **Azure OpenAI** model. \n- Uses a defined **prompt** to instruct the LLM on how to write the description. \n- Produces a clear, engaging product description that can be tailored to brand or SEO needs. \n- Allows easy replacement of Azure OpenAI with any other LLM node in n8n. "
},
"typeVersion": 1
},
{
"id": "3eef5219-71a7-4409-a135-658c8fe42a5c",
"name": "Sticky Note2",
"type": "n8n-nodes-base.stickyNote",
"position": [
-208,
-336
],
"parameters": {
"color": 2,
"width": 560,
"height": 704,
"content": "### Step 3: Save Product in Magento\n- Takes the generated description and updates the product via the **Magento 2 API**. \n- Ensures the new description is stored and visible on the webshop. \n- Supports updating other fields if desired (e.g. short description, meta description). \n- Confirms the product is enriched without requiring manual editing. "
},
"typeVersion": 1
},
{
"id": "62670ee5-58ad-45fc-b07e-f6d0885c4b98",
"name": "Sticky Note3",
"type": "n8n-nodes-base.stickyNote",
"position": [
-2464,
-336
],
"parameters": {
"width": 880,
"height": 1664,
"content": "# \ud83d\udecd\ufe0f Automated Product Description Generation for Adobe Commerce (Magento 2)\n\n## Description\nThis n8n template demonstrates how to automatically generate **product descriptions** for items in **Adobe Commerce (Magento 2)** that are missing one. \n\nThe workflow retrieves product data, converts raw attribute values (like numeric IDs) into **human-readable labels**, and passes the enriched product data to an **LLM** (Azure OpenAI by default). The LLM generates a compelling description, which is then saved back to Magento using the API. \n\nThis ensures all products have professional descriptions without manual writing effort.\n\n**Use cases include:** \n- Auto-generating missing descriptions for catalog completeness. \n- Creating consistent descriptions across large product datasets. \n- Reducing manual workload for content managers. \n- Tailoring descriptions for SEO and customer readability. \n\n---\n\n## Good to know\n- All **attribute options** are resolved to human-readable labels before being sent to the LLM. \n- The flow uses **Azure OpenAI**, but you can replace it with **OpenAI, Anthropic, Gemini, or other LLM providers**. \n- The LLM prompt can be **customised** to adjust tone, length, SEO-focus, or specific brand style. \n- Works out-of-the-box with Adobe Commerce (Magento 2) APIs, but can be adapted for other ecommerce systems. \n\n---\n\n## How it works\n1. **Get Product from Magento** \n - Retrieves a product that has no description. \n - Collects all product attributes.\n\n2. **Generate Description with LLM** \n - Resolves attribute option IDs into **human-readable values** (e.g. `color_id = 23 \u2192 \"Red\"`). \n - Passes the readable product attributes to an **Azure OpenAI model**. \n - The LLM creates a clear, engaging product description. \n - The prompt can be customised (e.g. SEO-optimized, short catalog text, or marketing style). \n\n3. **Save Description in Magento** \n - Updates the product via the **Magento API** with the generated description. \n - Ensures product data is enriched and visible in the webshop immediately. \n\n---\n\n## How to use\n- Configure your **Magento 2 API credentials** in n8n. \n- Replace the **Azure OpenAI** node with another provider if needed. \n- Adjust the **prompt** to match your brand\u2019s tone of voice. \n- Run the workflow to automatically process products missing descriptions. \n\n---\n\n## Requirements\n- \u2705 n8n instance (self-hosted or cloud) \n- \u2705 Adobe Commerce (Magento 2) instance with API access \n- \u2705 Azure OpenAI (or other LLM provider) credentials \n- (Optional) Prompt customisations for SEO or brand voice \n\n---\n\n## Customising this workflow\nThis workflow can be adapted for: \n- **Other attributes**: Include or exclude attributes (e.g. only color & size for apparel). \n- **Different LLMs**: Swap Azure OpenAI for OpenAI, Anthropic, Gemini, or any supported n8n AI node. \n- **Prompt tuning**: Adjust instructions to generate shorter, longer, or SEO-rich descriptions. \n- **Selective updates**: Target only specific categories (e.g. electronics, fashion). \n- **Multi-language support**: Generate product descriptions in multiple languages for international shops. \n\n---\n"
},
"typeVersion": 1
}
],
"active": false,
"settings": {
"executionOrder": "v1"
},
"versionId": "11259216-9fa3-4fea-b872-e36c88d412c8",
"connections": {
"Basic LLM Chain": {
"main": [
[
{
"node": "set Description",
"type": "main",
"index": 0
}
]
]
},
"set Description": {
"main": [
[
{
"node": "Save configurable",
"type": "main",
"index": 0
}
]
]
},
"Save configurable": {
"main": [
[],
[]
]
},
"Prepare attributes": {
"main": [
[
{
"node": "Basic LLM Chain",
"type": "main",
"index": 0
}
]
]
},
"Wait until nodes ready": {
"main": [
[
{
"node": "Prepare attributes",
"type": "main",
"index": 0
}
]
]
},
"Azure OpenAI Chat Model": {
"ai_languageModel": [
[
{
"node": "Basic LLM Chain",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Get attributes and options": {
"main": [
[
{
"node": "Wait until nodes ready",
"type": "main",
"index": 1
}
]
]
},
"get Product without description": {
"main": [
[
{
"node": "Wait until nodes ready",
"type": "main",
"index": 0
}
]
]
},
"When clicking \u2018Test workflow\u2019": {
"main": [
[
{
"node": "Get attributes and options",
"type": "main",
"index": 0
},
{
"node": "get Product without description",
"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.
azureOpenAiApimagento2Api
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
This n8n template demonstrates how to automatically generate product descriptions for items in Adobe Commerce (Magento 2) that are missing one.
Source: https://n8n.io/workflows/8475/ — 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.
My workflow 53. Uses formTrigger, httpRequest, lmChatOpenAi, form. Event-driven trigger; 74 nodes.
Form Trigger, HTTP Request, OpenAI Chat +15
Episode 23: UGC with nanobanana. Uses lmChatOpenAi, lmChatOllama, lmChatDeepSeek, lmChatOpenRouter. Event-driven trigger; 74 nodes.
OpenAI Chat, Ollama Chat, Lm Chat Deep Seek +12
Ultimate Blogblizt is a powerhouse workflow that solves the tedious task of crafting and publishing SEO-optimized tech blog posts. It integrates AI models (OpenAI, Google Gemini), WordPress, and multi
Chain Llm, Telegram Trigger, OpenAI Chat +10
Awesome N8N Templates. Uses agent, telegramTrigger, mcpClientTool, mcpTrigger. Event-driven trigger; 33 nodes.
Agent, Telegram Trigger, Mcp Client Tool +9
This workflow turns your sales calls + intake form into a polished, send-ready proposal. It pulls the latest call transcript from Fireflies, generates structured proposal content with Azure OpenAI, bu
Form Trigger, Chain Llm, Lm Chat Azure Open Ai +5