Alex Huy This workflow corresponds to n8n.io template #4476 — we link there as the canonical source.
This workflow follows the Agent → OpenAI Embeddings 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": "6bdbfe7d-5763-4047-876a-c61aae517887",
"name": "AI Agent",
"type": "@n8n/n8n-nodes-langchain.agent",
"notes": "\ud83e\udd16 CORE AI AGENT\n\nThis is the main LangChain agent that:\n- Receives formatted conversation history\n- Uses vector database for context retrieval\n- Can search web when needed\n- Maintains conversation memory\n- Generates intelligent responses\n\nConnected Tools:\n\u2705 Vector Store (knowledge base)\n\u2705 SerpAPI (web search)\n\u2705 Memory buffer\n\u2705 OpenAI GPT-4",
"position": [
720,
420
],
"parameters": {},
"typeVersion": 1.7
},
{
"id": "07c33b29-da37-4283-a9b9-314ffd3247d0",
"name": "Extract Message",
"type": "n8n-nodes-base.set",
"notes": "\ud83d\udcdd MESSAGE EXTRACTION\n\nExtracts key data from Google Sheets:\n- user_message: The actual question/request\n- chat_id: Unique identifier for this conversation\n\nThis data flows to:\n\u2192 Chat history retrieval\n\u2192 Response formatting",
"position": [
-80,
420
],
"parameters": {},
"typeVersion": 3.4
},
{
"id": "7c33317a-95a5-4f21-92b3-7c1dcf97bc43",
"name": "Format Conversation",
"type": "n8n-nodes-base.function",
"notes": "\ud83d\udd04 CONVERSATION FORMATTER\n\nProcesses chat history into proper format:\n\n1. Takes chat history from Google Sheets\n2. Converts to OpenAI message format:\n - role: 'system' | 'user' | 'assistant'\n - content: message text\n3. Keeps last 10 messages for context\n4. Adds current user message\n5. Includes system prompt with instructions\n\nOutput: Properly formatted conversation array for AI agent",
"position": [
360,
420
],
"parameters": {},
"typeVersion": 1
},
{
"id": "bfcd6c1b-2db4-49c0-9855-c8d5f3e979ab",
"name": "OpenAI Chat Model",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"notes": "\ud83e\udde0 MAIN LANGUAGE MODEL\n\nOpenAI GPT-4.1-nano for:\n- Natural conversation\n- Complex reasoning\n- Task completion\n- Creative responses\n\nSetup Required:\n\u2192 Add OpenAI API credentials\n\u2192 Ensure sufficient quota",
"position": [
580,
640
],
"parameters": {},
"typeVersion": 1
},
{
"id": "eadfe308-8c1d-4a41-a218-a091c53df572",
"name": "Embeddings OpenAI",
"type": "@n8n/n8n-nodes-langchain.embeddingsOpenAi",
"notes": "\ud83c\udfaf VECTOR EMBEDDINGS\n\nConverts text to numerical vectors for:\n- Semantic similarity search\n- Context retrieval from knowledge base\n- Intelligent document matching\n\nUses OpenAI's text-embedding-3-small model",
"position": [
800,
1040
],
"parameters": {},
"typeVersion": 1
},
{
"id": "141cb0cc-f004-495f-9ce9-2261daca70d7",
"name": "Vector Store LLM",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"notes": "\ud83d\udd0d KNOWLEDGE BASE LLM\n\nDedicated model for vector store queries:\n- Processes retrieved documents\n- Summarizes relevant context\n- Filters information for main agent\n\nSeparate from main model for efficiency",
"position": [
1000,
840
],
"parameters": {},
"typeVersion": 1
},
{
"id": "d48725fd-6f18-4ea6-8d26-4a2b9252c317",
"name": "Vector Store Tool",
"type": "@n8n/n8n-nodes-langchain.toolVectorStore",
"notes": "\ud83d\uddc3\ufe0f KNOWLEDGE BASE TOOL\n\nSearches vector database for relevant context:\n- Retrieves top 8 most similar documents\n- Uses semantic similarity matching\n- Provides context to main AI agent\n\nConnected to Supabase vector store",
"position": [
820,
640
],
"parameters": {},
"typeVersion": 1
},
{
"id": "d8859552-a2d7-4ada-a57e-2d542f80cbf1",
"name": "Supabase Vector Store",
"type": "@n8n/n8n-nodes-langchain.vectorStoreSupabase",
"notes": "\ud83d\udcbe VECTOR DATABASE\n\nSupabase vector store setup:\n\n1. Create 'documents' table in Supabase\n2. Enable vector extension (pgvector)\n3. Store embeddings + metadata\n4. Enable RLS policies\n\nTable columns needed:\n- id, content, metadata, embedding",
"position": [
720,
840
],
"parameters": {},
"typeVersion": 1
},
{
"id": "544ca207-6c7d-4f70-85bc-e9954fb404a4",
"name": "Conversation Memory",
"type": "@n8n/n8n-nodes-langchain.memoryBufferWindow",
"notes": "\ud83e\udde0 CONVERSATION MEMORY\n\nMaintains context across interactions:\n- Remembers recent conversation\n- Uses custom session key\n- Provides continuity for AI agent\n\nMemory Type: Buffer Window\n(keeps recent messages in memory)",
"position": [
700,
640
],
"parameters": {},
"typeVersion": 1.3
},
{
"id": "1c4c501e-6d24-43ab-9781-1ca2b36898a1",
"name": "Google Sheets Trigger",
"type": "n8n-nodes-base.googleSheetsTrigger",
"notes": "\ud83d\ude80 WORKFLOW TRIGGER\n\nMonitors Google Sheets for new messages:\n\nRequired Columns:\n- user_message (trigger column)\n- ai_respond (AI response)\n- id (unique identifier)\n\nSetup Steps:\n1. Create Google Sheet with columns\n2. Share with service account\n3. Connect credentials\n4. Set polling interval (every minute)\n\nTriggers when user_message column changes",
"position": [
-520,
250
],
"parameters": {},
"typeVersion": 1
},
{
"id": "6f3c6a51-12d3-41af-8c2f-e84a7d442f80",
"name": "Check If New Message",
"type": "n8n-nodes-base.if",
"notes": "\u2705 DUPLICATE PREVENTION\n\nPrevents processing already-answered messages:\n\n\u2713 If ai_respond is empty \u2192 Process message\n\u2717 If ai_respond has content \u2192 Skip\n\nThis ensures each user message is only processed once, avoiding infinite loops or duplicate responses.",
"position": [
-300,
250
],
"parameters": {},
"typeVersion": 2.2
},
{
"id": "7019c655-2740-420c-8fe6-b1342ae3dd13",
"name": "Merge Response Data",
"type": "n8n-nodes-base.merge",
"notes": "\ud83d\udd17 DATA MERGER\n\nCombines data for final update:\n\nInput 1: Original message + timestamp ID\nInput 2: AI agent response\n\nOutput: Complete record ready for Google Sheets update with all required fields",
"position": [
1120,
200
],
"parameters": {},
"typeVersion": 3.1
},
{
"id": "47034580-420c-4b11-870e-07be8ea37436",
"name": "Generate Timestamp ID",
"type": "n8n-nodes-base.function",
"notes": "\ud83d\udd50 UNIQUE ID GENERATOR\n\nGenerates unique timestamp-based ID:\n- Uses current timestamp (milliseconds)\n- Ensures each message has unique identifier\n- Prevents duplicate entries\n- Used for Google Sheets row identification\n\nField: myNewField \u2192 ID column",
"position": [
-20,
180
],
"parameters": {},
"typeVersion": 1
},
{
"id": "575fc087-be86-40c1-b7d2-7d8e0d9fb7f5",
"name": "Get Chat History",
"type": "n8n-nodes-base.googleSheets",
"notes": "\ud83d\udcda CHAT HISTORY RETRIEVAL\n\nRetrieves previous conversations:\n- Reads all rows from Google Sheets\n- Provides context for conversation formatting\n- Enables memory across sessions\n\nUsed by Format Conversation node to build proper chat history for AI agent",
"position": [
140,
420
],
"parameters": {},
"typeVersion": 4.6
},
{
"id": "7b4a3d27-03aa-4229-90b6-ba5180af0eea",
"name": "Update AI Response",
"type": "n8n-nodes-base.googleSheets",
"notes": "\ud83d\udcbe RESPONSE STORAGE\n\nUpdates Google Sheets with AI response:\n\nOperation: appendOrUpdate\n- Matches by user_message column\n- Updates ai_respond with agent output\n- Adds unique ID for tracking\n\nThis completes the conversation loop and makes the response visible to users",
"position": [
1440,
360
],
"parameters": {},
"typeVersion": 4.6
},
{
"id": "892ac40d-b3dc-4cb5-b634-0afae0e91a71",
"name": "Web Search Tool",
"type": "@n8n/n8n-nodes-langchain.toolSerpApi",
"notes": "\ud83c\udf10 WEB SEARCH CAPABILITY\n\nProvides real-time web search for AI agent:\n- Access to current information\n- Google search results\n- Supplements knowledge base\n\nSetup Required:\n\u2192 SerpAPI account & API key\n\u2192 Sufficient search quota\n\nAI agent automatically uses when needed",
"position": [
1120,
640
],
"parameters": {},
"typeVersion": 1
},
{
"id": "8169a7cc-1585-44d7-80c2-e5ebdfd37532",
"name": "Sticky Note",
"type": "n8n-nodes-base.stickyNote",
"position": [
-580,
140
],
"parameters": {
"content": ""
},
"typeVersion": 1
},
{
"id": "32cb9197-527d-4d12-b67a-51cecf2b866e",
"name": "Sticky Note7",
"type": "n8n-nodes-base.stickyNote",
"position": [
-1240,
140
],
"parameters": {
"content": ""
},
"typeVersion": 1
},
{
"id": "0d2f99a7-8140-4679-802d-65c0336e351a",
"name": "Sticky Note6",
"type": "n8n-nodes-base.stickyNote",
"position": [
-1240,
580
],
"parameters": {
"content": ""
},
"typeVersion": 1
}
],
"connections": {
"AI Agent": {
"main": [
[
{
"node": "Merge Response Data",
"type": "main",
"index": 1
}
]
]
},
"Extract Message": {
"main": [
[
{
"node": "Get Chat History",
"type": "main",
"index": 0
}
]
]
},
"Web Search Tool": {
"ai_tool": [
[
{
"node": "AI Agent",
"type": "ai_tool",
"index": 0
}
]
]
},
"Get Chat History": {
"main": [
[
{
"node": "Format Conversation",
"type": "main",
"index": 0
}
]
]
},
"Vector Store LLM": {
"ai_languageModel": [
[
{
"node": "Vector Store Tool",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Embeddings OpenAI": {
"ai_embedding": [
[
{
"node": "Supabase Vector Store",
"type": "ai_embedding",
"index": 0
}
]
]
},
"OpenAI Chat Model": {
"ai_languageModel": [
[
{
"node": "AI Agent",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Vector Store Tool": {
"ai_tool": [
[
{
"node": "AI Agent",
"type": "ai_tool",
"index": 0
}
]
]
},
"Conversation Memory": {
"ai_memory": [
[
{
"node": "AI Agent",
"type": "ai_memory",
"index": 0
}
]
]
},
"Format Conversation": {
"main": [
[
{
"node": "AI Agent",
"type": "main",
"index": 0
}
]
]
},
"Merge Response Data": {
"main": [
[
{
"node": "Update AI Response",
"type": "main",
"index": 0
}
]
]
},
"Check If New Message": {
"main": [
[
{
"node": "Extract Message",
"type": "main",
"index": 0
},
{
"node": "Generate Timestamp ID",
"type": "main",
"index": 0
}
]
]
},
"Generate Timestamp ID": {
"main": [
[
{
"node": "Merge Response Data",
"type": "main",
"index": 0
}
]
]
},
"Google Sheets Trigger": {
"main": [
[
{
"node": "Check If New Message",
"type": "main",
"index": 0
}
]
]
},
"Supabase Vector Store": {
"ai_vectorStore": [
[
{
"node": "Vector Store Tool",
"type": "ai_vectorStore",
"index": 0
}
]
]
}
}
}
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
Description An intelligent conversational AI system that provides contextual responses by combining chat history, vector database knowledge retrieval, and web search capabilities. How it Works (High-level steps)
Source: https://n8n.io/workflows/4476/ — 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.
Who is this for? This workflow is ideal for HR teams, startups, and enterprises that want to handle employee interactions through WhatsApp and automate responses using LLM (OpenAI) and intelligent rou
WhatsApp Trigger, OpenAI, OpenAI Chat +13
BambooHR AI-Powered Company Policies and Benefits Chatbot. Uses manualTrigger, documentDefaultDataLoader, embeddingsOpenAi, textSplitterRecursiveCharacterTextSplitter. Event-driven trigger; 50 nodes.
Document Default Data Loader, OpenAI Embeddings, Text Splitter Recursive Character Text Splitter +14
BambooHR AI-Powered Company Policies and Benefits Chatbot. Uses manualTrigger, documentDefaultDataLoader, embeddingsOpenAi, textSplitterRecursiveCharacterTextSplitter. Event-driven trigger; 50 nodes.
Document Default Data Loader, OpenAI Embeddings, Text Splitter Recursive Character Text Splitter +14
This workflow enables companies to provide instant HR support by automating responses to employee queries about policies and benefits: Retrieves company policies, benefits, and HR documents from Bambo
Document Default Data Loader, OpenAI Embeddings, Text Splitter Recursive Character Text Splitter +14
Turn your docs into an AI-powered internal or public-facing assistant. This chatbot workflow uses RAG (Retrieval-Augmented Generation) with Supabase vector search to answer employee or customer questi
Supabase Vector Store, Document Default Data Loader, Text Splitter Recursive Character Text Splitter +11