This workflow follows the HTTP Request → OpenAI 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 →
{
"name": "Ask AI About Document (RAG)",
"nodes": [
{
"parameters": {
"httpMethod": "POST",
"path": "ask-document",
"responseMode": "responseNode",
"options": {}
},
"id": "webhook-trigger",
"name": "Webhook Trigger",
"type": "n8n-nodes-base.webhook",
"typeVersion": 1,
"position": [
240,
300
]
},
{
"parameters": {
"conditions": {
"options": {
"caseSensitive": true,
"leftValue": "",
"typeValidation": "strict"
},
"conditions": [
{
"id": "document-id-check",
"leftValue": "={{ $json.document_id }}",
"rightValue": "",
"operator": {
"type": "string",
"operation": "notEmpty"
}
},
{
"id": "question-check",
"leftValue": "={{ $json.question }}",
"rightValue": "",
"operator": {
"type": "string",
"operation": "notEmpty"
}
}
],
"combinator": "and"
},
"options": {}
},
"id": "validate-input",
"name": "Validate Input",
"type": "n8n-nodes-base.if",
"typeVersion": 2,
"position": [
460,
300
]
},
{
"parameters": {
"model": "text-embedding-3-small",
"inputType": "text",
"input": "={{ $json.question }}",
"options": {}
},
"id": "generate-query-embedding",
"name": "Generate Query Embedding",
"type": "n8n-nodes-base.openAi",
"typeVersion": 1,
"position": [
680,
200
]
},
{
"parameters": {
"url": "https://YOUR_SUPABASE_PROJECT.supabase.co/rest/v1/rpc/match_document_chunks",
"authentication": "predefinedCredentialType",
"nodeCredentialType": "supabaseApi",
"sendHeaders": true,
"headerParameters": {
"parameters": [
{
"name": "apikey",
"value": "={{ $credentials.supabaseApi.serviceRole }}"
},
{
"name": "Authorization",
"value": "Bearer {{ $credentials.supabaseApi.serviceRole }}"
},
{
"name": "Content-Type",
"value": "application/json"
}
]
},
"sendBody": true,
"specifyBody": "json",
"jsonBody": "={\n \"query_embedding\": {{ JSON.stringify($json.embedding) }},\n \"match_threshold\": 0.7,\n \"match_count\": 5,\n \"doc_id\": \"{{ $('Webhook Trigger').first().json.document_id }}\"\n}",
"options": {}
},
"id": "vector-search",
"name": "Vector Search",
"type": "n8n-nodes-base.httpRequest",
"typeVersion": 4.2,
"position": [
900,
200
]
},
{
"parameters": {
"url": "https://YOUR_SUPABASE_PROJECT.supabase.co/rest/v1/documents",
"authentication": "predefinedCredentialType",
"nodeCredentialType": "supabaseApi",
"sendHeaders": true,
"headerParameters": {
"parameters": [
{
"name": "apikey",
"value": "={{ $credentials.supabaseApi.serviceRole }}"
},
{
"name": "Authorization",
"value": "Bearer {{ $credentials.supabaseApi.serviceRole }}"
}
]
},
"sendQuery": true,
"queryParameters": {
"parameters": [
{
"name": "id",
"value": "eq.{{ $('Webhook Trigger').first().json.document_id }}"
},
{
"name": "select",
"value": "title,document_type,ai_summary"
}
]
},
"options": {}
},
"id": "get-document-info",
"name": "Get Document Info",
"type": "n8n-nodes-base.httpRequest",
"typeVersion": 4.2,
"position": [
900,
400
]
},
{
"parameters": {
"jsCode": "// Build context from search results\nconst searchResults = $('Vector Search').first().json;\nconst documentInfo = $('Get Document Info').first().json[0];\nconst question = $('Webhook Trigger').first().json.question;\n\n// Extract relevant chunks\nconst contextChunks = searchResults.map(result => ({\n text: result.chunk_text,\n similarity: result.similarity,\n chunk_index: result.chunk_index\n}));\n\n// Build context string\nconst context = contextChunks\n .map((chunk, index) => `[Chunk ${chunk.chunk_index}]: ${chunk.text}`)\n .join('\\n\\n');\n\n// Create system prompt\nconst systemPrompt = `You are an AI assistant helping users understand documents. You have access to relevant excerpts from a document titled \"${documentInfo.title}\" (type: ${documentInfo.document_type}).\n\nDocument Summary: ${documentInfo.ai_summary || 'No summary available'}\n\nInstructions:\n1. Answer the user's question based ONLY on the provided context\n2. If the context doesn't contain enough information, say so clearly\n3. Cite specific chunks when possible (e.g., \"According to Chunk 2...\")\n4. Be concise but comprehensive\n5. If asked about something not in the context, explain that you can only answer based on the provided document excerpts`;\n\nconst userPrompt = `Context from document:\\n${context}\\n\\nQuestion: ${question}\\n\\nPlease provide a helpful answer based on the context above.`;\n\nreturn [{\n json: {\n system_prompt: systemPrompt,\n user_prompt: userPrompt,\n context_chunks: contextChunks.map(c => c.chunk_index),\n question: question,\n document_id: $('Webhook Trigger').first().json.document_id,\n user_id: $('Webhook Trigger').first().json.user_id\n }\n}];"
},
"id": "build-context",
"name": "Build Context",
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
1120,
300
]
},
{
"parameters": {
"model": "gpt-4o-mini",
"messages": {
"messageValues": [
{
"role": "system",
"message": "={{ $json.system_prompt }}"
},
{
"role": "user",
"message": "={{ $json.user_prompt }}"
}
]
},
"options": {
"temperature": 0.3,
"maxTokens": 500
}
},
"id": "generate-answer",
"name": "Generate Answer",
"type": "n8n-nodes-base.openAi",
"typeVersion": 1,
"position": [
1340,
300
]
},
{
"parameters": {
"url": "https://YOUR_SUPABASE_PROJECT.supabase.co/rest/v1/document_chats",
"authentication": "predefinedCredentialType",
"nodeCredentialType": "supabaseApi",
"sendHeaders": true,
"headerParameters": {
"parameters": [
{
"name": "apikey",
"value": "={{ $credentials.supabaseApi.serviceRole }}"
},
{
"name": "Authorization",
"value": "Bearer {{ $credentials.supabaseApi.serviceRole }}"
},
{
"name": "Content-Type",
"value": "application/json"
}
]
},
"sendBody": true,
"specifyBody": "json",
"jsonBody": "={\n \"document_id\": \"{{ $('Build Context').first().json.document_id }}\",\n \"user_id\": \"{{ $('Build Context').first().json.user_id }}\",\n \"question\": \"{{ $('Build Context').first().json.question }}\",\n \"answer\": \"{{ $json.message.content }}\",\n \"context_chunks\": {{ JSON.stringify($('Build Context').first().json.context_chunks) }}\n}",
"options": {}
},
"id": "save-chat",
"name": "Save Chat",
"type": "n8n-nodes-base.httpRequest",
"typeVersion": 4.2,
"position": [
1560,
300
]
},
{
"parameters": {
"respondWith": "json",
"responseBody": "={\n \"success\": true,\n \"answer\": \"{{ $('Generate Answer').first().json.message.content }}\",\n \"context_chunks\": {{ JSON.stringify($('Build Context').first().json.context_chunks) }},\n \"document_id\": \"{{ $('Build Context').first().json.document_id }}\",\n \"question\": \"{{ $('Build Context').first().json.question }}\",\n \"timestamp\": \"{{ new Date().toISOString() }}\"\n}"
},
"id": "success-response",
"name": "Success Response",
"type": "n8n-nodes-base.respondToWebhook",
"typeVersion": 1,
"position": [
1780,
300
]
},
{
"parameters": {
"respondWith": "json",
"responseBody": "={\n \"success\": false,\n \"error\": \"Invalid input: document_id and question are required\",\n \"received\": {{ JSON.stringify($json) }}\n}",
"options": {
"responseCode": 400
}
},
"id": "error-response",
"name": "Error Response",
"type": "n8n-nodes-base.respondToWebhook",
"typeVersion": 1,
"position": [
680,
400
]
},
{
"parameters": {
"conditions": {
"options": {
"caseSensitive": true,
"leftValue": "",
"typeValidation": "strict"
},
"conditions": [
{
"id": "has-results",
"leftValue": "={{ $json.length }}",
"rightValue": 0,
"operator": {
"type": "number",
"operation": "gt"
}
}
],
"combinator": "and"
},
"options": {}
},
"id": "check-search-results",
"name": "Check Search Results",
"type": "n8n-nodes-base.if",
"typeVersion": 2,
"position": [
1120,
200
]
},
{
"parameters": {
"respondWith": "json",
"responseBody": "={\n \"success\": true,\n \"answer\": \"I couldn't find relevant information in this document to answer your question. The document may not contain the information you're looking for, or your question might be too specific. Try rephrasing your question or asking about the general content of the document.\",\n \"context_chunks\": [],\n \"document_id\": \"{{ $('Webhook Trigger').first().json.document_id }}\",\n \"question\": \"{{ $('Webhook Trigger').first().json.question }}\",\n \"timestamp\": \"{{ new Date().toISOString() }}\",\n \"note\": \"No relevant context found\"\n}"
},
"id": "no-context-response",
"name": "No Context Response",
"type": "n8n-nodes-base.respondToWebhook",
"typeVersion": 1,
"position": [
1340,
100
]
}
],
"connections": {
"Webhook Trigger": {
"main": [
[
{
"node": "Validate Input",
"type": "main",
"index": 0
}
]
]
},
"Validate Input": {
"main": [
[
{
"node": "Generate Query Embedding",
"type": "main",
"index": 0
},
{
"node": "Get Document Info",
"type": "main",
"index": 0
}
],
[
{
"node": "Error Response",
"type": "main",
"index": 0
}
]
]
},
"Generate Query Embedding": {
"main": [
[
{
"node": "Vector Search",
"type": "main",
"index": 0
}
]
]
},
"Vector Search": {
"main": [
[
{
"node": "Check Search Results",
"type": "main",
"index": 0
}
]
]
},
"Get Document Info": {
"main": [
[
{
"node": "Build Context",
"type": "main",
"index": 0
}
]
]
},
"Build Context": {
"main": [
[
{
"node": "Generate Answer",
"type": "main",
"index": 0
}
]
]
},
"Generate Answer": {
"main": [
[
{
"node": "Save Chat",
"type": "main",
"index": 0
}
]
]
},
"Save Chat": {
"main": [
[
{
"node": "Success Response",
"type": "main",
"index": 0
}
]
]
},
"Check Search Results": {
"main": [
[
{
"node": "Build Context",
"type": "main",
"index": 0
}
],
[
{
"node": "No Context Response",
"type": "main",
"index": 0
}
]
]
}
},
"active": false,
"settings": {
"executionOrder": "v1"
},
"versionId": "1",
"meta": {
"templateCredsSetupCompleted": true
},
"id": "ask-ai-rag",
"tags": []
}
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
Ask AI About Document (RAG). Uses openAi, httpRequest. Webhook trigger; 12 nodes.
Source: https://github.com/madhavchaturvedi005/KMRL-MODEL/blob/bc8f98043f0a5b04ba1607cccdd0b276d856016b/n8n-workflows/02-ask-ai-rag.json — 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 captures new leads via webhook, enriches and scores them with Apollo and Google Gemini, logs everything in Google Sheets, and automates outreach, reply handling, follow-ups, meeting prep
HTTP Request, Google Gemini, Google Sheets +3
This powerful n8n automation workflow is designed to execute advanced B2B lead enrichment and hyper-personalization for cold email outreach. By orchestrating a complex chain of data scraping, AI analy
OpenAI, HTTP Request, Airtable
Propulsar — Content Engine v3. Uses openAi, httpRequest, googleSheets. Webhook trigger; 73 nodes.
OpenAI, HTTP Request, Google Sheets
Eu Clara – Funil Kiwify Completo. Uses postgres, openAi, httpRequest, gmail. Webhook trigger; 70 nodes.
Postgres, OpenAI, HTTP Request +1
Postagem Redes — Portal: Ações. Uses openAi, googleGemini, ollama, httpRequest. Webhook trigger; 58 nodes.
OpenAI, Google Gemini, Ollama +4