Salim BRAHMI This workflow corresponds to n8n.io template #17762 — we link there as the canonical source.
This workflow follows the Agent → Documentdefaultdataloader 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": "9Z7ucLPwzGsulHly",
"meta": {
"templateCredsSetupCompleted": true
},
"name": "Product Backlog RAG Assistant",
"tags": [
{
"id": "RKP1E9UWmf8t5UuL",
"name": "Draft",
"createdAt": "2026-07-01T15:26:33.507Z",
"updatedAt": "2026-07-01T15:26:33.507Z"
}
],
"nodes": [
{
"id": "cb59bb61-dda4-4f5e-817e-98360c670425",
"name": "Sticky Note",
"type": "n8n-nodes-base.stickyNote",
"position": [
-1120,
1360
],
"parameters": {
"width": 480,
"height": 800,
"content": "## Product Backlog RAG Assistant\n\n### How it works\n\n1. A form submission triggers extraction of file content and stores it in a Supabase vector store.\n2. Default data and embeddings are prepared to initialize the knowledge base.\n3. A Slack slash command triggers an immediate acknowledgment and initiates a RAG agent to answer queries.\n4. The RAG agent uses embeddings to search the vector store and generate a response via an AI model.\n5. The answer is posted back to Slack via an HTTP request.\n\n### Setup steps\n\n- - [ ] Configure the form trigger (e.g., n8n form or webhook) to capture file uploads.\n- - [ ] Set up Supabase credentials for the vector store node.\n- - [ ] Configure Google Gemini API key for both embeddings nodes.\n- - [ ] Set up Slack credentials for the slash command webhook and outgoing message.\n- - [ ] Optionally configure the AI model (OpenRouter) used in the RAG agent.\n\n### Customization\n\nYou can modify the form fields to capture different metadata, switch to another vector store like Pinecone, adjust the prompt in the RAG agent for answer style, or change the Slack response format."
},
"typeVersion": 1
},
{
"id": "198c996f-69e1-4e77-9957-3d967351adc1",
"name": "Sticky Note1",
"type": "n8n-nodes-base.stickyNote",
"position": [
-544,
1360
],
"parameters": {
"color": 7,
"width": 1024,
"height": 272,
"content": "## Form-based Data Ingestion\n\nExtracts file content from form submissions and stores it in a Supabase vector store."
},
"typeVersion": 1
},
{
"id": "4fb98b89-25d1-4f11-a44c-52e17c812dae",
"name": "Sticky Note2",
"type": "n8n-nodes-base.stickyNote",
"position": [
32,
1664
],
"parameters": {
"color": 7,
"width": 448,
"height": 304,
"content": "## Default Knowledge Base Setup\n\nLoads default data and generates embeddings to initialize the vector store."
},
"typeVersion": 1
},
{
"id": "42d74ffd-0857-4557-b338-e16adb09915d",
"name": "Sticky Note3",
"type": "n8n-nodes-base.stickyNote",
"position": [
-560,
1696
],
"parameters": {
"color": 7,
"width": 464,
"height": 464,
"content": "## Slack Command Acknowledgment\n\nReceives Slack slash commands and sends an immediate acknowledgment response."
},
"typeVersion": 1
},
{
"id": "34097603-df2e-4a23-89f7-9a197262d6cd",
"name": "Sticky Note4",
"type": "n8n-nodes-base.stickyNote",
"position": [
-64,
2000
],
"parameters": {
"color": 7,
"width": 624,
"height": 720,
"content": "## RAG Agent Processing\n\nGenerates query embeddings and runs the AI agent to search the knowledge base."
},
"typeVersion": 1
},
{
"id": "88156866-a547-4be2-9ec5-5abf941f04b2",
"name": "Sticky Note5",
"type": "n8n-nodes-base.stickyNote",
"position": [
624,
1936
],
"parameters": {
"color": 7,
"width": 272,
"height": 368,
"content": "## Answer Posting to Slack\n\nPosts the generated answer back to Slack via an HTTP request."
},
"typeVersion": 1
},
{
"id": "8f7589c1-e586-43dd-94e5-05660ad604d8",
"name": "On form submission",
"type": "n8n-nodes-base.formTrigger",
"position": [
-496,
1472
],
"parameters": {
"options": {},
"formTitle": "Index Product Backlog",
"formFields": {
"values": [
{
"fieldType": "file",
"fieldLabel": "Backlog PDF",
"acceptFileTypes": ".pdf"
}
]
},
"formDescription": "Upload your product backlog PDF to make it searchable via Slack."
},
"typeVersion": 2.5
},
{
"id": "7b092acb-ffe1-44d2-855a-e995fd5ecaa2",
"name": "Extract from File",
"type": "n8n-nodes-base.extractFromFile",
"position": [
-176,
1472
],
"parameters": {
"options": {},
"operation": "pdf",
"binaryPropertyName": "Backlog_PDF"
},
"typeVersion": 1.1
},
{
"id": "01072a7d-6450-4b7d-a006-e286ce421f52",
"name": "Supabase Vector Store",
"type": "@n8n/n8n-nodes-langchain.vectorStoreSupabase",
"position": [
160,
1472
],
"parameters": {
"mode": "insert",
"options": {},
"tableName": {
"__rl": true,
"mode": "list",
"value": "backlog_documents",
"cachedResultName": "backlog_documents"
},
"embeddingBatchSize": "=200"
},
"credentials": {
"supabaseApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.3
},
{
"id": "48d3b927-65c1-49ba-ba88-774987721a2e",
"name": "Default Data Loader",
"type": "@n8n/n8n-nodes-langchain.documentDefaultDataLoader",
"position": [
320,
1792
],
"parameters": {
"options": {}
},
"typeVersion": 1.1
},
{
"id": "2f6e1e57-769c-4aac-9824-c8853b23da8c",
"name": "Embeddings Google Gemini",
"type": "@n8n/n8n-nodes-langchain.embeddingsGoogleGemini",
"position": [
112,
1792
],
"parameters": {
"modelName": "models/gemini-embedding-001"
},
"credentials": {
"googlePalmApi": {
"name": "<your credential>"
}
},
"typeVersion": 1
},
{
"id": "6725f522-3fab-4d1f-9000-02ff18b01bb0",
"name": "Slack Slash Command Trigger",
"type": "n8n-nodes-base.webhook",
"position": [
-512,
1984
],
"parameters": {
"path": "backlog-query",
"options": {},
"httpMethod": "POST",
"responseMode": "responseNode"
},
"typeVersion": 2.1
},
{
"id": "0646ffd6-ec93-4a8c-896d-11a3c9bbdcbf",
"name": "Respond to Webhook",
"type": "n8n-nodes-base.respondToWebhook",
"position": [
-288,
1840
],
"parameters": {
"options": {},
"respondWith": "json",
"responseBody": "={\n \"response_type\": \"in_channel\",\n \"text\": \"\ud83d\udd0d *Question backlog :* {{ $('Slack Slash Command Trigger').item.json.body.text }}\\n_Recherche en cours..._\"\n}"
},
"typeVersion": 1.5
},
{
"id": "99083ddb-836f-4cdd-be74-4f6a42bef71c",
"name": "OpenRouter Chat Model",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenRouter",
"position": [
-16,
2384
],
"parameters": {
"model": "openrouter/free",
"options": {
"temperature": 0.8
}
},
"credentials": {
"openRouterApi": {
"name": "<your credential>"
}
},
"typeVersion": 1
},
{
"id": "68d3925f-3768-4d33-b59d-9448738873ae",
"name": "search_backlog",
"type": "@n8n/n8n-nodes-langchain.vectorStoreSupabase",
"position": [
304,
2368
],
"parameters": {
"mode": "retrieve-as-tool",
"topK": 5,
"options": {},
"tableName": {
"__rl": true,
"mode": "list",
"value": "backlog_documents",
"cachedResultName": "backlog_documents"
},
"toolDescription": "Search the product backlog for user stories, epics, acceptance criteria, and requirements. Use this to answer any question about the backlog content."
},
"credentials": {
"supabaseApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.3
},
{
"id": "96917c05-b096-40bb-baff-0c7d95c109db",
"name": "Embeddings Google Gemini1",
"type": "@n8n/n8n-nodes-langchain.embeddingsGoogleGemini",
"position": [
304,
2544
],
"parameters": {
"modelName": "models/gemini-embedding-001"
},
"credentials": {
"googlePalmApi": {
"name": "<your credential>"
}
},
"typeVersion": 1
},
{
"id": "c59dc945-1123-4b38-ba2c-78d255f27236",
"name": "RAG Agent",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
160,
2128
],
"parameters": {
"text": "={{ $('Slack Slash Command Trigger').item.json.body.text }}",
"options": {
"systemMessage": "You are a product backlog assistant. You answer questions about a product backlog by searching it with the \"search_backlog\" tool.\n\nRules:\n- ALWAYS use the search_backlog tool before answering. Never answer from memory.\n- Base your answer ONLY on the retrieved backlog content. Do not invent stories, \n IDs, or requirements that aren't in the search results.\n- If the backlog doesn't contain the answer, say so clearly: \"Je ne trouve pas \n cette information dans le backlog.\"\n- When you cite a user story, include its ID (e.g. NTS-1) so the user can find it.\n- Answer in the same language as the question.\n- Be concise and structured. Use bullet points for lists of stories.\n- For \"how many\" or \"list all\" questions, search broadly and enumerate what you find.\n\nFormat your answer for Slack: use *bold* (single asterisks), bullet points \nwith \u2022, and keep it scannable."
},
"promptType": "define"
},
"typeVersion": 3.1
},
{
"id": "351a927c-20ab-44c2-906d-f4e796fc9c76",
"name": "Post Answer to Slack",
"type": "n8n-nodes-base.httpRequest",
"position": [
704,
2128
],
"parameters": {
"url": "={{ $('Slack Slash Command Trigger').item.json.body.response_url }}",
"method": "POST",
"options": {},
"jsonBody": "={{ JSON.stringify({\n response_type: \"in_channel\",\n replace_original: true,\n blocks: [\n {\n type: \"section\",\n text: {\n type: \"mrkdwn\",\n text: `\ud83d\udd0d *Question :* ${$('Slack Slash Command Trigger').item.json.body.text}`\n }\n },\n { type: \"divider\" },\n {\n type: \"section\",\n text: {\n type: \"mrkdwn\",\n text: ($json.output && $json.output.trim().length > 0)\n ? $json.output\n : \"_Je n'ai pas pu g\u00e9n\u00e9rer de r\u00e9ponse. V\u00e9rifie que le backlog est bien index\u00e9._\"\n }\n }\n ]\n}) }}",
"sendBody": true,
"specifyBody": "json"
},
"typeVersion": 4.4
}
],
"active": false,
"settings": {
"binaryMode": "separate",
"executionOrder": "v1"
},
"versionId": "eb73d118-6af1-4a21-9550-668968d7ae60",
"connections": {
"RAG Agent": {
"main": [
[
{
"node": "Post Answer to Slack",
"type": "main",
"index": 0
}
]
]
},
"search_backlog": {
"ai_tool": [
[
{
"node": "RAG Agent",
"type": "ai_tool",
"index": 0
}
]
]
},
"Extract from File": {
"main": [
[
{
"node": "Supabase Vector Store",
"type": "main",
"index": 0
}
]
]
},
"On form submission": {
"main": [
[
{
"node": "Extract from File",
"type": "main",
"index": 0
}
]
]
},
"Default Data Loader": {
"ai_document": [
[
{
"node": "Supabase Vector Store",
"type": "ai_document",
"index": 0
}
]
]
},
"OpenRouter Chat Model": {
"ai_languageModel": [
[
{
"node": "RAG Agent",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Embeddings Google Gemini": {
"ai_embedding": [
[
{
"node": "Supabase Vector Store",
"type": "ai_embedding",
"index": 0
}
]
]
},
"Embeddings Google Gemini1": {
"ai_embedding": [
[
{
"node": "search_backlog",
"type": "ai_embedding",
"index": 0
}
]
]
},
"Slack Slash Command Trigger": {
"main": [
[
{
"node": "Respond to Webhook",
"type": "main",
"index": 0
},
{
"node": "RAG Agent",
"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.
googlePalmApiopenRouterApisupabaseApi
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
This workflow indexes a product backlog PDF into a Supabase vector store and lets users query it from Slack using a slash command, with Google Gemini embeddings for retrieval and an OpenRouter chat model to generate answers. Receives a form submission with a product backlog PDF…
Source: https://n8n.io/workflows/17762/ — 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 intelligent chatbot leverages cutting-edge financial APIs and AI-driven analysis to deliver comprehensive stock research reports. Get instant access to professional-grade investment analysis that
Tool Think, Supabase Vector Store, OpenAI Embeddings +15
This workflow ingests incident and playbook content from GitHub into Supabase (including pgvector embeddings with Google Gemini) and, on a webhook trigger, enriches a test incident with historical mat
Document Default Data Loader, Google Gemini Embeddings, HTTP Request +9
A production-ready 3-workflow system that handles customer support across WhatsApp and Email using RAG-powered AI. Automatically routes queries, detects escalation intent, logs handoffs to Google Shee
Form Trigger, Pinecone Vector Store, Document Default Data Loader +12
Automatically ingest your knowledge base, monitor your LinkedIn company page inbox, and reply to leads and customers with AI-drafted, source-grounded answers. Classifies every message, checks for spam
Google Drive Trigger, Google Drive, Pinecone Vector Store +9
My workflow 2. Uses googleGemini, formTrigger, httpRequest, googleDrive. Event-driven trigger; 33 nodes.
Google Gemini, Form Trigger, HTTP Request +8