This workflow follows the Agent → Anthropic 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 →
{
"nodes": [
{
"id": "1205b121-8aaa-4e41-874b-4e81aad6374e",
"name": "Sticky Note",
"type": "n8n-nodes-base.stickyNote",
"position": [
120,
600
],
"parameters": {
"color": 4,
"width": 462.4041757955455,
"height": 315.6388466176832,
"content": "## Generate SEO Seed Keywords Using AI\n\nThis flow uses an AI node to generate Seed Keywords to focus SEO efforts on based on your ideal customer profile\n\n**Outputs:** \n- List of 20 Seed Keywords\n\n\n**Pre-requisites / Dependencies:**\n- You know your ideal customer profile (ICP)\n- An AI API account (either OpenAI or Anthropic recommended)"
},
"typeVersion": 1
},
{
"id": "d2654d75-2b64-4ec3-b583-57d2b6b7b195",
"name": "Sticky Note13",
"type": "n8n-nodes-base.stickyNote",
"disabled": true,
"position": [
640,
920
],
"parameters": {
"color": 7,
"width": 287.0816455493243,
"height": 330.47923074942287,
"content": "**Generate draft seed KW based on ICP**\n\n"
},
"typeVersion": 1
},
{
"id": "d248a58e-3705-4b6f-99cb-e9187e56781c",
"name": "Anthropic Chat Model",
"type": "@n8n/n8n-nodes-langchain.lmChatAnthropic",
"position": [
680,
1120
],
"parameters": {
"options": {}
},
"typeVersion": 1.2
},
{
"id": "71517d83-59f5-441a-8a75-c35f4e06a8a2",
"name": "Split Out",
"type": "n8n-nodes-base.splitOut",
"position": [
980,
980
],
"parameters": {
"options": {},
"fieldToSplitOut": "output.answer"
},
"typeVersion": 1
},
{
"id": "1c68eff5-6478-4eba-9abe-3ccea2a17a5c",
"name": "Sticky Note12",
"type": "n8n-nodes-base.stickyNote",
"disabled": true,
"position": [
120,
920
],
"parameters": {
"color": 7,
"width": 492.16246201447336,
"height": 213.62075341687063,
"content": "**Get data from airtable and format** "
},
"typeVersion": 1
},
{
"id": "53dcc524-ef3d-40b8-b79d-976517dce4e7",
"name": "Sticky Note17",
"type": "n8n-nodes-base.stickyNote",
"disabled": true,
"position": [
960,
920
],
"parameters": {
"color": 7,
"width": 348.42891651921957,
"height": 213.62075341687063,
"content": "**Add data to database**"
},
"typeVersion": 1
},
{
"id": "570495fe-3f1d-44ae-bea0-9fa4b2ce15ef",
"name": "Sticky Note11",
"type": "n8n-nodes-base.stickyNote",
"position": [
640,
820
],
"parameters": {
"color": 6,
"width": 393.46745700785266,
"height": 80,
"content": "**Costs to run**\nApprox. $0.02-0.05 for a run using Claude Sonnet 3.5"
},
"typeVersion": 1
},
{
"id": "6e5e84c5-409f-4f37-931a-21a44aff7c5e",
"name": "Set Ideal Customer Profile (ICP)",
"type": "n8n-nodes-base.set",
"position": [
160,
980
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "973e949e-1afd-4378-8482-d2168532eff6",
"name": "product",
"type": "string",
"value": "=**Replace this with a string detailing your intended product (if you have one)**"
},
{
"id": "ce9c0a8f-6157-4b46-8b77-133545dc71bd",
"name": "pain points",
"type": "string",
"value": "=**Replace this with a string list of customer pain points**"
},
{
"id": "5abc858a-c412-4acf-acb9-488e4d992d2f",
"name": "goals",
"type": "string",
"value": "=**Replace this with a string list of your customers key goals/objectives**"
},
{
"id": "fbdd1ef7-c1b9-48eb-b73e-a383f12b5ba1",
"name": "current solutions",
"type": "string",
"value": "=**Replace this with a string detailing how your ideal customer currently solves their pain ppoints**"
},
{
"id": "2e5c8f48-266e-486c-956f-51f1449f6288",
"name": "expertise level",
"type": "string",
"value": "=**Replace this with a string detailing customer level of expertise**"
}
]
}
},
"notesInFlow": true,
"typeVersion": 3.4
},
{
"id": "bd5781f4-6f35-45d3-8182-12ea6712eddf",
"name": "Aggregate for AI node",
"type": "n8n-nodes-base.aggregate",
"position": [
380,
980
],
"parameters": {
"options": {},
"aggregate": "aggregateAllItemData"
},
"notesInFlow": true,
"typeVersion": 1
},
{
"id": "244943bf-e4dd-40fc-9a43-7a5cd0da1c5b",
"name": "Sticky Note14",
"type": "n8n-nodes-base.stickyNote",
"position": [
640,
1260
],
"parameters": {
"color": 3,
"width": 284.87764467541297,
"height": 80,
"content": "**REQUIRED**\nConnect to your own AI API above"
},
"typeVersion": 1
},
{
"id": "73c8f47a-4fdb-40c8-9062-890ef1265ab0",
"name": "Sticky Note16",
"type": "n8n-nodes-base.stickyNote",
"position": [
120,
1140
],
"parameters": {
"color": 3,
"width": 284.87764467541297,
"height": 80,
"content": "**REQUIRED**\nSet your Ideal Customer Profile before proceeding"
},
"typeVersion": 1
},
{
"id": "a5b93e6d-44ab-4b6f-b86a-25dc621b52b0",
"name": "AI Agent",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
660,
980
],
"parameters": {
"text": "=User:\nHere are some important rules for you to follow:\n<rules>\n1. Analyze the ICP information carefully.\n2. Generate 15-20 seed keywords that are relevant to the ICP's needs, challenges, goals, and search behavior.\n3. Ensure the keywords are broad enough to be considered \"\"head\"\" terms, but specific enough to target the ICP effectively.\n4. Consider various aspects of the ICP's journey, including awareness, consideration, and decision stages.\n5. Include a mix of product-related, problem-related, and solution-related terms.\n6. Think beyond just the product itself - consider industry trends, related technologies, and broader business concepts that would interest the ICP.\n7. Avoid overly generic terms that might attract irrelevant traffic.\n8. Aim for a mix of keyword difficulties, including both competitive and less competitive terms.\n9. Include keywords that cover different search intents: informational, navigational, commercial, and transactional.\n10. Consider related tools or platforms that the ICP might use, and include relevant integration-related keywords.\n11. If applicable, include some location-specific keywords based on the ICP's geographic information.\n12. Incorporate industry-specific terminology or jargon that the ICP would likely use in their searches.\n13. Consider emerging trends or pain points in the ICP's industry that they might be searching for solutions to.\n13. Format the keywords in lowercase, without punctuation. Trim any leading or trailing white space.\n</rules>\n\nYour output should be an array of strings, each representing a seed keyword:\n<example>\n['b2b lead generation', 'startup marketing strategies', 'saas sales funnel', ...]\n</example>\n\nHere is the Ideal Customer Profile (ICP) information:\n<input>\n{{ $json.data[0].product }}\n</input>\n\nNow:\nBased on the provided ICP, generate an array of 15-20 seed keywords that will form the foundation of a comprehensive SEO strategy for this B2B SaaS company. These keywords should reflect a deep understanding of the ICP's needs, challenges, and search behavior, while also considering broader industry trends and related concepts.\n\nFirst, write out your ideas in {thoughts: } JSON as part of your analysis, then answer inside the {answer: } key in the JSON. ",
"agent": "conversationalAgent",
"options": {
"systemMessage": "=System: You are an expert SEO strategist tasked with generating 15-20 key head search terms (seed keywords) for a B2B SaaS company. Your goal is to create a comprehensive list of keywords that will attract and engage the ideal customer profile (ICP) described."
},
"promptType": "define"
},
"typeVersion": 1.6
},
{
"id": "ca3c0bd5-7ef0-4e2b-9b5e-071773c32c85",
"name": "Connect to your own database",
"type": "n8n-nodes-base.noOp",
"position": [
1140,
980
],
"parameters": {},
"typeVersion": 1
},
{
"id": "94639a81-5e46-482a-851a-5443bfe9863c",
"name": "Sticky Note15",
"type": "n8n-nodes-base.stickyNote",
"position": [
1120,
1140
],
"parameters": {
"color": 3,
"width": 284.87764467541297,
"height": 80,
"content": "**REQUIRED**\nConnect to your own database / GSheet / Airtable base to output these"
},
"typeVersion": 1
},
{
"id": "16498e92-c0d5-44f4-b993-c9c8930955bc",
"name": "When clicking \u2018Test workflow\u2019",
"type": "n8n-nodes-base.manualTrigger",
"position": [
-60,
980
],
"parameters": {},
"typeVersion": 1
}
],
"connections": {
"AI Agent": {
"main": [
[
{
"node": "Split Out",
"type": "main",
"index": 0
}
]
]
},
"Split Out": {
"main": [
[
{
"node": "Connect to your own database",
"type": "main",
"index": 0
}
]
]
},
"Anthropic Chat Model": {
"ai_languageModel": [
[
{
"node": "AI Agent",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Aggregate for AI node": {
"main": [
[
{
"node": "AI Agent",
"type": "main",
"index": 0
}
]
]
},
"Set Ideal Customer Profile (ICP)": {
"main": [
[
{
"node": "Aggregate for AI node",
"type": "main",
"index": 0
}
]
]
},
"When clicking \u2018Test workflow\u2019": {
"main": [
[
{
"node": "Set Ideal Customer Profile (ICP)",
"type": "main",
"index": 0
}
]
]
}
}
}
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
How this works
Business owners and marketers can quickly generate a comprehensive list of SEO seed keywords tailored to their niche, saving hours of manual brainstorming and ensuring relevance from the start. This workflow leverages AI to transform a basic topic or product description into targeted keyword ideas, ideal for content creators, e-commerce managers, or digital agencies aiming to boost search visibility without deep technical expertise. The key step involves prompting the Anthropic Chat Model with your input, which intelligently expands it into seed keywords before an agent refines and prioritises them for maximum impact.
Use this workflow when launching a new website, refreshing content strategy, or exploring keyword opportunities for a specific audience, particularly if you're comfortable with event-driven triggers for on-demand execution. Avoid it for highly specialised industries requiring custom data sources, or when you need real-time competitor analysis beyond basic AI generation. Common variations include integrating additional nodes for keyword volume checks via Google Search Console or exporting results directly to Google Sheets for team collaboration.
About this workflow
Generate Seo Seed Keywords Using Ai. Uses stickyNote, lmChatAnthropic, splitOut, agent. Event-driven trigger; 15 nodes.
Source: https://github.com/Zie619/n8n-workflows — 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.
This workflow contains community nodes that are only compatible with the self-hosted version of n8n.
Personal Assistant. Uses memoryBufferWindow, agent, agentTool, httpRequestTool. Event-driven trigger; 77 nodes.
This template and YouTube video goes over 5 different implementations of evaluations within n8n. Categorization Correctness Tools used String similarity Helpfulness
SMS Outreach Engine: Service providers. Uses agent, lmChatAnthropic, httpRequest, googleSheets. Event-driven trigger; 60 nodes.
This workflow is designed for marketers, founders, agencies, and product teams who want to understand how real customers talk about a product category, market, or problem space.