n8nauto
Get Started

RAG & GenAI App With WordPress Content

Build a powerful RAG (Retrieval Augmented Generation) and Generative AI application using your WordPress content with this n8n workflow. This comprehensive solution automatically retrieves, processes, and stores your WordPress posts and pages, making them available for AI-powered queries. It connects WordPress to a PostgreSQL database (often hosted on Supabase) for vector storage and chat memory, leveraging OpenAI for embeddings and chat model interactions. This workflow is ideal for businesses and content creators who want to transform their existing WordPress knowledge base into an interactive AI assistant, customer support chatbot, or internal knowledge retrieval system, allowing users to ask natural language questions and receive accurate, context-aware answers directly from their published content. By automating the content synchronization and AI integration, it significantly reduces the manual effort of maintaining an AI knowledge base, ensures your AI always has access to the latest information, and enhances user engagement by providing instant access to information.

53 nodesmanual trigger104 views0 copiesMarketing
WordPressPostgreSQLSupabaseOpenAI

Workflow JSON

{"id": "o8iTqIh2sVvnuWz5", "meta": {"instanceId": "b9faf72fe0d7c3be94b3ebff0778790b50b135c336412d28fd4fca2cbbf8d1f5"}, "name": "RAG & GenAI App With WordPress Content", "tags": [], "nodes": [{"id": "c3738490-ed39-4774-b337-bf5ee99d0c72", "name": "When clicking \u2018Test workflow\u2019", "type": "n8n-nodes-base.manualTrigger", "position": [500, 940], "parameters": {}, "typeVersion": 1}, {"id": "3ab719bd-3652-433f-a597-9cd28f8cfcea", "name": "Embeddings OpenAI", "type": "@n8n/n8n-nodes-langchain.embeddingsOpenAi", "position": [2580, 1320], "parameters": {"model": "text-embedding-3-small", "options": {}}, "typeVersion": 1}, {"id": "e8639569-2091-44de-a84d-c3fc3ce54de4", "name": "Default Data Loader", "type": "@n8n/n8n-nodes-langchain.documentDefaultDataLoader", "position": [2800, 1260], "parameters": {"options": {"metadata": {"metadataValues": [{"name": "title", "value": "={{ $json.title }}"}, {"name": "url", "value": "={{ $json.url }}"}, {"name": "content_type", "value": "={{ $json.content_type }}"}, {"name": "publication_date", "value": "={{ $json.publication_date }}"}, {"name": "modification_date", "value": "={{ $json.modification_date }}"}, {"name": "id", "value": "={{ $json.id }}"}]}}, "jsonData": "={{ $json.data }}", "jsonMode": "expressionData"}, "typeVersion": 1}, {"id": "e7f858eb-4dca-40ea-9da9-af953687e63d", "name": "Token Splitter", "type": "@n8n/n8n-nodes-langchain.textSplitterTokenSplitter", "position": [2900, 1480], "parameters": {"chunkSize": 300, "chunkOverlap": 30}, "typeVersion": 1}, {"id": "27585104-5315-4c11-b333-4b5d27d9bae4", "name": "Embeddings OpenAI1", "type": "@n8n/n8n-nodes-langchain.embeddingsOpenAi", "position": [1400, 2340], "parameters": {"model": "text-embedding-3-small", "options": {}}, "typeVersion": 1}, {"id": "35269a98-d905-4e4f-ae5b-dadad678f260", "name": "OpenAI Chat Model", "type": "@n8n/n8n-nodes-langchain.lmChatOpenAi", "position": [2800, 2300], "parameters": {"model": "gpt-4o-mini", "options": {}}, "typeVersion": 1}, {"id": "cd26b6fa-a8bb-4139-9bec-8656d90d8203", "name": "Postgres Chat Memory", "type": "@n8n/n8n-nodes-langchain.memoryPostgresChat", "position": [2920, 2300], "parameters": {"tableName": "website_chat_histories"}, "typeVersion": 1.1}, {"id": "7c718e1b-1398-49f3-ba67-f970a82983e0", "name": "Respond to Webhook", "type": "n8n-nodes-base.respondToWebhook", "position": [3380, 2060], "parameters": {"options": {}}, "typeVersion": 1.1}, {"id": "f91f18e0-7a04-4218-8490-bff35dfbf7a8", "name": "Set fields", "type": "n8n-nodes-base.set", "position": [2360, 2060], "parameters": {"options": {}, "assignments": {"assignments": [{"id": "6888175b-853b-457a-96f7-33dfe952a05d", "name": "documents", "type": "string", "value": "={{ \n JSON.stringify(\n $json.documents.map(doc => ({\n metadata: \n 'URL: ' + doc.metadata.url.replaceAll('&rsquo;', \"'\").replaceAll(/[\"]/g, '') + '\\n' +\n 'Publication Date: ' + doc.metadata.publication_date.replaceAll(/[\"]/g, '') + '\\n' +\n 'Modification Date: ' + doc.metadata.modification_date.replaceAll(/[\"]/g, '') + '\\n' +\n 'Content Type: ' + doc.metadata.content_type.replaceAll(/[\"]/g, '') + '\\n' +\n 'Title: ' + doc.metadata.title.replaceAll('&rsquo;', \"'\").replaceAll(/[\"]/g, '') + '\\n',\n \n page_content: doc.pageContent\n }))\n ).replaceAll(/[\\[\\]{}]/g, '')\n}}"}, {"id": "ae310b77-4560-4f44-8c4e-8d13f680072e", "name": "sessionId", "type": "string", "value": "={{ $('When chat message received').item.json.sessionId }}"}, {"id": "8738f4de-b3c3-45ad-af4b-8311c8105c35", "name": "chatInput", "type": "string", "value": "={{ $('When chat message received').item.json.chatInput }}"}]}}, "typeVersion": 3.4}, {"id": "7f392a40-e353-4bb2-9ecf-3ee330110b95", "name": "Embeddings OpenAI2", "type": "@n8n/n8n-nodes-langchain.embeddingsOpenAi", "position": [6400, 860], "parameters": {"model": "text-embedding-3-small", "options": {}}, "typeVersion": 1}, {"id": "9e045857-5fcd-4c4b-83ee-ceda28195b76", "name": "Default Data Loader1", "type": "@n8n/n8n-nodes-langchain.documentDefaultDataLoader", "position": [6500, 860], "parameters": {"options": {"metadata": {"metadataValues": [{"name": "title", "value": "={{ $json.title }}"}, {"name": "url", "value": "={{ $json.url }}"}, {"name": "content_type", "value": "={{ $json.content_type }}"}, {"name": "publication_date", "value": "={{ $json.publication_date }}"}, {"name": "modification_date", "value": "={{ $json.modification_date }}"}, {"name": "id", "value": "={{ $json.id }}"}]}}, "jsonData": "={{ $json.data }}", "jsonMode": "expressionData"}, "typeVersion": 1}, {"id": "d0c1144b-4542-470e-8cbe-f985e839d9d0", "name": "Token Splitter1", "type": "@n8n/n8n-nodes-langchain.textSplitterTokenSplitter", "position": [6500, 980], "parameters": {"chunkSize": 300, "chunkOverlap": 30}, "typeVersion": 1}, {"id": "ec7cf1b2-f56f-45da-bb34-1dc8a66a7de6", "name": "Markdown1", "type": "n8n-nodes-base.markdown", "position": [6240, 900], "parameters": {"html": "={{ $json.content }}", "options": {}}, "typeVersion": 1}, {"id": "8399976b-340a-49ce-a5b6-f7339957aa9d", "name": "Postgres", "type": "n8n-nodes-base.postgres", "position": [4260, 900], "parameters": {"query": "select max(created_at) as last_workflow_execution from n8n_website_embedding_histories", "options": {}, "operation": "executeQuery"}, "typeVersion": 2.5}, {"id": "88e79403-06df-4f18-9e4c-a4c4e727aa17", "name": "Aggregate", "type": "n8n-nodes-base.aggregate", "position": [3300, 900], "parameters": {"options": {}, "aggregate": "aggregateAllItemData"}, "typeVersion": 1}, {"id": "db7241e8-1c3a-4f91-99b7-383000f41afe", "name": "Aggregate1", "type": "n8n-nodes-base.aggregate", "position": [6800, 680], "parameters": {"options": {}, "aggregate": "aggregateAllItemData"}, "typeVersion": 1}, {"id": "94bbba31-d83b-427f-a7dc-336725238294", "name": "Aggregate2", "type": "n8n-nodes-base.aggregate", "position": [7180, 1160], "parameters": {"options": {}, "fieldsToAggregate": {"fieldToAggregate": [{"fieldToAggregate": "metadata.id"}]}}, "typeVersion": 1}, {"id": "52a110fa-cdd6-4b1d-99fe-394b5dfa0a1f", "name": "Sticky Note", "type": "n8n-nodes-base.stickyNote", "position": [440, 600], "parameters": {"color": 5, "width": 3308.2687575224263, "height": 1015.3571428571431, "content": "# Workflow 1 : Initial Embedding \n## Use this workflow to create the initial embedding for your WordPress website content\n\n"}, "typeVersion": 1}, {"id": "4cbf8135-a52b-4a54-b7b0-15ea27ce7ae3", "name": "Sticky Note1", "type": "n8n-nodes-base.stickyNote", "position": [3812, 605], "parameters": {"color": 5, "width": 3785.6673412474183, "height": 1020.4528919414245, "content": "# Workflow 2 : Upsert\n## Use this workflow to upsert embeddings for documents stored in the Supabase vector table\n"}, "typeVersion": 1}, {"id": "f6e954e0-a37a-45ac-9882-20f4f1944b70", "name": "Sticky Note2", "type": "n8n-nodes-base.stickyNote", "position": [440, 1820], "parameters": {"color": 5, "width": 3235.199999999999, "height": 817.9199999999992, "content": "# Workflow 3 : Use this workflow to enable chat functionality with your website content. The chat can be embedded into your website to enhance user experience"}, "typeVersion": 1}, {"id": "acbdd54b-f02a-41aa-a0ce-8642db560151", "name": "Wordpress - Get all posts", "type": "n8n-nodes-base.wordpress", "position": [1260, 880], "parameters": {"options": {}, "operation": "getAll", "returnAll": true}, "typeVersion": 1}, {"id": "94fce59d-9336-4d49-a378-17335ec02e52", "name": "Wordpress - Get all pages", "type": "n8n-nodes-base.wordpress", "position": [1260, 1060], "parameters": {"options": {}, "resource": "page", "operation": "getAll", "returnAll": true}, "typeVersion": 1}, {"id": "b00c92e5-1765-4fd9-9981-e01053992a0a", "name": "Sticky Note3", "type": "n8n-nodes-base.stickyNote", "position": [1157, 727], "parameters": {"width": 1108.3519999999999, "height": 561.4080000000004, "content": "## Use filters to create embeddings only for content that you want to include in your GenAI application"}, "typeVersion": 1}, {"id": "f8a22739-898d-456b-93f8-79f74b60a00c", "name": "Set fields1", "type": "n8n-nodes-base.set", "position": [2320, 900], "parameters": {"options": {}, "assignments": {"assignments": [{"id": "de6711dc-d03c-488c-bef4-0a853e2d0a14", "name": "publication_date", "type": "string", "value": "={{ $json.date }}"}, {"id": "f8e35dcc-c96c-4554-b6bc-8e5d7eca90e3", "name": "modification_date", "type": "string", "value": "={{ $json.modified }}"}, {"id": "f6a6e3de-fe39-4cfc-ab07-c4ccfaef78f5", "name": "content_type", "type": "string", "value": "={{ $json.type }}"}, {"id": "b0428598-073f-4560-9a0c-01caf3708921", "name": "title", "type": "string", "value": "={{ $json.title.rendered }}"}, {"id": "534f51b4-b43a-40d3-8120-58df8043d909", "name": "url", "type": "string", "value": "={{ $json.link }}"}, {"id": "dbe0c559-90bd-49f8-960e-0d85d5ed4f5e", "name": "content", "type": "string", "value": "={{ $json.content.rendered }}"}, {"id": "892be7c6-b032-4129-b285-1986ed4ee046", "name": "protected", "type": "boolean", "value": "={{ $json.excerpt.protected }}"}, {"id": "06fac885-4431-41ff-a43b-6eb84ca57401", "name": "status", "type": "string", "value": "={{ $json.status }}"}, {"id": "43b1aea7-895e-41da-a0a6-2f1cec1f1b97", "name": "id", "type": "number", "value": "={{ $json.id }}"}]}}, "typeVersion": 3.4}, {"id": "404db031-f470-4e42-a3b3-66b849a86174", "name": "Filter - Only published & unprotected content", "type": "n8n-nodes-base.filter", "position": [2520, 900], "parameters": {"options": {}, "conditions": {"options": {"version": 2, "leftValue": "", "caseSensitive": true, "typeValidation": "strict"}, "combinator": "and", "conditions": [{"id": "1f708587-f3d3-487a-843a-b6a2bfad2ca9", "operator": {"type": "boolean", "operation": "false", "singleValue": true}, "leftValue": "={{ $json.protected }}", "rightValue": ""}, {"id": "04f47269-e112-44c3-9014-749898aca8bd", "operator": {"name": "filter.operator.equals", "type": "string", "operation": "equals"}, "leftValue": "={{ $json.status }}", "rightValue": "publish"}]}}, "typeVersion": 2.2}, {"id": "05bb6091-515e-4f22-a3fd-d25b2046a03d", "name": "HTML To Markdown", "type": "n8n-nodes-base.markdown", "position": [2740, 900], "parameters": {"html": "={{ $json.content}}", "options": {}}, "typeVersion": 1}, {"id": "391e9ea7-71dd-42ae-bee7-badcae32427c", "name": "Supabase - Store workflow execution", "type": "n8n-nodes-base.supabase", "position": [3520, 900], "parameters": {"tableId": "n8n_website_embedding_histories", "fieldsUi": {"fieldValues": [{"fieldId": "id", "fieldValue": "={{ $executionId }}"}]}}, "typeVersion": 1}, {"id": "47dad096-efc8-4bdd-9c22-49562325d8a0", "name": "Sticky Note4", "type": "n8n-nodes-base.stickyNote", "position": [460, 1320], "parameters": {"width": 851.1898437499999, "height": 275.2000000000001, "content": "## Run these two nodes if the \"documents\" table on Supabase and the \"n8n_website_embedding_histories\" table do not exist"}, "typeVersion": 1}, {"id": "d19f3a5f-fa42-46d0-a366-4c5a5d09f559", "name": "Every 30 seconds", "type": "n8n-nodes-base.scheduleTrigger", "position": [3940, 900], "parameters": {"rule": {"interval": [{"field": "seconds"}]}}, "typeVersion": 1.2}, {"id": "a22ab0dd-1da8-4fc2-8106-6130bf7938c8", "name": "Sticky Note5", "type": "n8n-nodes-base.stickyNote", "position": [3820, 740], "parameters": {"width": 336.25, "height": 292.5, "content": "## Set this node to match the frequency of publishing and updating on your website"}, "typeVersion": 1}, {"id": "ba25135b-6e6e-406b-b18a-f532a6e37276", "name": "Wordpress - Get posts modified after last workflow execution", "type": "n8n-nodes-base.httpRequest", "position": [4600, 840], "parameters": {"url": "https://mydomain.com/wp-json/wp/v2/posts", "options": {}, "sendQuery": true, "authentication": "predefinedCredentialType", "queryParameters": {"parameters": [{"name": "modified_after", "value": "={{ $json.last_workflow_execution }}"}]}, "nodeCredentialType": "wordpressApi"}, "typeVersion": 4.2}, {"id": "a1d8572e-2b0d-40a1-a898-bbd563a6b190", "name": "Wordpress - Get posts modified after last workflow execution1", "type": "n8n-nodes-base.httpRequest", "position": [4600, 1060], "parameters": {"url": "https://mydomain.com/wp-json/wp/v2/pages", "options": {}, "sendQuery": true, "authentication": "predefinedCredentialType", "queryParameters": {"parameters": [{"name": "modified_after", "value": "={{ $json.last_workflow_execution }}"}]}, "nodeCredentialType": "wordpressApi"}, "typeVersion": 4.2}, {"id": "c0839aaa-8ba7-47ff-8fa9-dc75e1c4da84", "name": "Set fields2", "type": "n8n-nodes-base.set", "position": [5420, 920], "parameters": {"options": {}, "assignments": {"assignments": [{"id": "de6711dc-d03c-488c-bef4-0a853e2d0a14", "name": "publication_date", "type": "string", "value": "={{ $json.date }}"}, {"id": "f8e35dcc-c96c-4554-b6bc-8e5d7eca90e3", "name": "modification_date", "type": "string", "value": "={{ $json.modified }}"}, {"id": "f6a6e3de-fe39-4cfc-ab07-c4ccfaef78f5", "name": "content_type", "type": "string", "value": "={{ $json.type }}"}, {"id": "b0428598-073f-4560-9a0c-01caf3708921", "name": "title", "type": "string", "value": "={{ $json.title.rendered }}"}, {"id": "534f51b4-b43a-40d3-8120-58df8043d909", "name": "url", "type": "string", "value": "={{ $json.link }}"}, {"id": "dbe0c559-90bd-49f8-960e-0d85d5ed4f5e", "name": "content", "type": "string", "value": "={{ $json.content.rendered }}"}, {"id": "892be7c6-b032-4129-b285-1986ed4ee046", "name": "protected", "type": "boolean", "value": "={{ $json.content.protected }}"}, {"id": "06fac885-4431-41ff-a43b-6eb84ca57401", "name": "status", "type": "string", "value": "={{ $json.status }}"}, {"id": "43b1aea7-895e-41da-a0a6-2f1cec1f1b97", "name": "id", "type": "number", "value": "={{ $json.id }}"}]}}, "typeVersion": 3.4}, {"id": "15b1d30a-5861-4380-89d5-0eef65240503", "name": "Filter - Only published and unprotected content", "type": "n8n-nodes-base.filter", "position": [5760, 920], "parameters": {"options": {}, "conditions": {"options": {"version": 2, "leftValue": "", "caseSensitive": true, "typeValidation": "strict"}, "combinator": "and", "conditions": [{"id": "c2b25d74-91d7-44ea-8598-422100947b07", "operator": {"type": "boolean", "operation": "false", "singleValue": true}, "leftValue": "={{ $json.protected }}", "rightValue": ""}, {"id": "3e63bf79-25ca-4ccf-aa86-ff5f90e1ece1", "operator": {"name": "filter.operator.equals", "type": "string", "operation": "equals"}, "leftValue": "={{ $json.status }}", "rightValue": "publish"}]}}, "typeVersion": 2.2}, {"id": "0990f503-8d6f-44f6-8d04-7e2f7d74301a", "name": "Loop Over Items", "type": "n8n-nodes-base.splitInBatches", "position": [6040, 920], "parameters": {"options": {}}, "typeVersion": 3}, {"id": "6cc4e46e-3884-4259-b7ed-51c5552cc3e0", "name": "Set fields3", "type": "n8n-nodes-base.set", "position": [7400, 1160], "parameters": {"options": {}, "assignments": {"assignments": [{"id": "de6711dc-d03c-488c-bef4-0a853e2d0a14", "name": "publication_date", "type": "string", "value": "={{ $('Loop Over Items').item.json.publication_date }}"}, {"id": "f8e35dcc-c96c-4554-b6bc-8e5d7eca90e3", "name": "modification_date", "type": "string", "value": "={{ $('Loop Over Items').item.json.modification_date }}"}, {"id": "f6a6e3de-fe39-4cfc-ab07-c4ccfaef78f5", "name": "content_type", "type": "string", "value": "={{ $('Loop Over Items').item.json.content_type }}"}, {"id": "b0428598-073f-4560-9a0c-01caf3708921", "name": "title", "type": "string", "value": "={{ $('Loop Over Items').item.json.title }}"}, {"id": "534f51b4-b43a-40d3-8120-58df8043d909", "name": "url", "type": "string", "value": "={{ $('Loop Over Items').item.json.url }}"}, {"id": "dbe0c559-90bd-49f8-960e-0d85d5ed4f5e", "name": "content", "type": "string", "value": "={{ $('Loop Over Items').item.json.content }}"}, {"id": "892be7c6-b032-4129-b285-1986ed4ee046", "name": "protected", "type": "boolean", "value": "={{ $('Loop Over Items').item.json.protected }}"}, {"id": "06fac885-4431-41ff-a43b-6eb84ca57401", "name": "status", "type": "string", "value": "={{ $('Loop Over Items').item.json.status }}"}, {"id": "43b1aea7-895e-41da-a0a6-2f1cec1f1b97", "name": "id", "type": "number", "value": "={{ $('Loop Over Items').item.json.id }}"}]}}, "typeVersion": 3.4}, {"id": "24f47982-a803-4848-8390-c400a8cebcee", "name": "Set fields4", "type": "n8n-nodes-base.set", "position": [6680, 1400], "parameters": {"options": {}, "assignments": {"assignments": [{"id": "de6711dc-d03c-488c-bef4-0a853e2d0a14", "name": "publication_date", "type": "string", "value": "={{ $('Loop Over Items').item.json.publication_date }}"}, {"id": "f8e35dcc-c96c-4554-b6bc-8e5d7eca90e3", "name": "modification_date", "type": "string", "value": "={{ $('Loop Over Items').item.json.modification_date }}"}, {"id": "f6a6e3de-fe39-4cfc-ab07-c4ccfaef78f5", "name": "content_type", "type": "string", "value": "={{ $('Loop Over Items').item.json.content_type }}"}, {"id": "b0428598-073f-4560-9a0c-01caf3708921", "name": "title", "type": "string", "value": "={{ $('Loop Over Items').item.json.title }}"}, {"id": "534f51b4-b43a-40d3-8120-58df8043d909", "name": "url", "type": "string", "value": "={{ $('Loop Over Items').item.json.url }}"}, {"id": "dbe0c559-90bd-49f8-960e-0d85d5ed4f5e", "name": "content", "type": "string", "value": "={{ $('Loop Over Items').item.json.content }}"}, {"id": "892be7c6-b032-4129-b285-1986ed4ee046", "name": "protected", "type": "boolean", "value": "={{ $('Loop Over Items').item.json.protected }}"}, {"id": "06fac885-4431-41ff-a43b-6eb84ca57401", "name": "status", "type": "string", "value": "={{ $('Loop Over Items').item.json.status }}"}, {"id": "43b1aea7-895e-41da-a0a6-2f1cec1f1b97", "name": "id", "type": "number", "value": "={{ $('Loop Over Items').item.json.id }}"}]}}, "typeVersion": 3.4}, {"id": "5f59ebbf-ca17-4311-809c-85b74ce624cc", "name": "Store documents on Supabase", "type": "@n8n/n8n-nodes-langchain.vectorStoreSupabase", "position": [6380, 680], "parameters": {"mode": "insert", "options": {"queryName": "match_documents"}, "tableName": {"__rl": true, "mode": "list", "value": "documents", "cachedResultName": "documents"}}, "typeVersion": 1}, {"id": "2422562e-9c95-4d77-ae8c-485b06f9234e", "name": "Store workflow execution id and timestamptz", "type": "n8n-nodes-base.supabase", "position": [7060, 680], "parameters": {"tableId": "n8n_website_embedding_histories"}, "typeVersion": 1}, {"id": "5013f3a1-f7fb-4fa7-9ef2-3599f77f5fc8", "name": "Aggregate documents", "type": "n8n-nodes-base.aggregate", "position": [1960, 2060], "parameters": {"options": {}, "fieldsToAggregate": {"fieldToAggregate": [{"renameField": true, "outputFieldName": "documents", "fieldToAggregate": "document"}]}}, "typeVersion": 1}, {"id": "26532217-3206-4be3-b186-733bc364913b", "name": "Sticky Note6", "type": "n8n-nodes-base.stickyNote", "position": [1220, 1980], "parameters": {"width": 665.78125, "height": 507.65625, "content": "## Retrieve documents from Supabase immediately after chat input to send metadata to OpenAI"}, "typeVersion": 1}, {"id": "78d2806c-8d13-44b8-bd6d-866fa794edae", "name": "Sticky Note7", "type": "n8n-nodes-base.stickyNote", "position": [6375, 1090], "parameters": {"width": 1198.9843749999998, "height": 515.4687499999998, "content": "## Switch:\n- **If the document exists and has been updated:** delete rows and insert new embedding\n- **If it\u2019s a new document:** insert embedding"}, "typeVersion": 1}, {"id": "3b5ffada-ae2a-45a2-a76c-69732b05761c", "name": "Postgres - Create documents table", "type": "n8n-nodes-base.postgres", "position": [560, 1440], "parameters": {"query": "-- Enable the pgvector extension to work with embedding vectors\nCREATE EXTENSION vector;\n\n-- Create a table to store your documents with default RLS\nCREATE TABLE\n documents (\n id BIGINT PRIMARY KEY GENERATED ALWAYS AS IDENTITY,\n CONTENT TEXT, -- corresponds to Document.pageContent\n metadata jsonb, -- corresponds to Document.metadata\n embedding vector (1536) -- 1536 works for OpenAI embeddings, change if needed\n );\n\n-- Enable Row Level Security on the documents table\nALTER TABLE documents ENABLE ROW LEVEL SECURITY;\n\n-- Create a function to search for documents\nCREATE FUNCTION match_documents (\n query_embedding vector (1536),\n match_count INT DEFAULT NULL,\n FILTER jsonb DEFAULT '{}'\n) RETURNS TABLE (\n id BIGINT,\n CONTENT TEXT,\n metadata jsonb,\n similarity FLOAT\n) LANGUAGE plpgsql AS $$\n#variable_conflict use_column\nBEGIN\n RETURN QUERY\n SELECT\n id,\n content,\n metadata,\n 1 - (documents.embedding <=> query_embedding) AS similarity\n FROM documents\n WHERE metadata @> filter\n ORDER BY documents.embedding <=> query_embedding\n LIMIT match_count;\nEND;\n$$;", "options": {}, "operation": "executeQuery"}, "typeVersion": 2.5}, {"id": "632a7b44-a062-472e-a777-805ee74a4bd6", "name": "Postgres - Create workflow execution history table", "type": "n8n-nodes-base.postgres", "position": [920, 1440], "parameters": {"query": "CREATE TABLE\n n8n_website_embedding_histories (\n id BIGINT PRIMARY KEY GENERATED ALWAYS AS IDENTITY,\n created_at TIMESTAMP WITH TIME ZONE DEFAULT NOW()\n );", "options": {}, "operation": "executeQuery"}, "typeVersion": 2.5}, {"id": "7c55e08b-e116-4e22-bd1d-e4bec5107d89", "name": "Merge Wordpress Posts and Pages", "type": "n8n-nodes-base.merge", "position": [1660, 900], "parameters": {}, "typeVersion": 3}, {"id": "4520db6c-2e68-45ff-9439-6fd95f95dc85", "name": "Merge retrieved WordPress posts and pages", "type": "n8n-nodes-base.merge", "position": [5120, 920], "parameters": {}, "typeVersion": 3}, {"id": "d547a063-6b76-4bfd-ba0a-165181c4af19", "name": "Postgres - Filter on existing documents", "type": "n8n-nodes-base.postgres", "position": [6260, 1180], "parameters": {"query": "SELECT *\nFROM documents\nWHERE (metadata->>'id')::integer = {{ $json.id }};\n", "options": {}, "operation": "executeQuery"}, "typeVersion": 2.5, "alwaysOutputData": true}, {"id": "03456a81-d512-4fd8-842a-27b6d8b3f94e", "name": "Supabase - Delete row if documents exists", "type": "n8n-nodes-base.supabase", "position": [6900, 1160], "parameters": {"tableId": "documents", "operation": "delete", "filterType": "string", "filterString": "=metadata->>id=like.{{ $json.metadata.id }}"}, "executeOnce": false, "typeVersion": 1, "alwaysOutputData": false}, {"id": "72e5bf4b-c413-4fb7-acb8-59e7abee60f7", "name": "Switch", "type": "n8n-nodes-base.switch", "position": [6580, 1180], "parameters": {"rules": {"values": [{"outputKey": "existing_documents", "conditions": {"options": {"version": 2, "leftValue": "", "caseSensitive": true, "typeValidation": "strict"}, "combinator": "and", "conditions": [{"operator": {"type": "number", "operation": "exists", "singleValue": true}, "leftValue": "={{ $json.metadata.id }}", "rightValue": ""}]}, "renameOutput": true}, {"outputKey": "new_documents", "conditions": {"options": {"version": 2, "leftValue": "", "caseSensitive": true, "typeValidation": "strict"}, "combinator": "and", "conditions": [{"id": "696d1c1b-8674-4549-880e-e0d0ff681905", "operator": {"type": "number", "operation": "notExists", "singleValue": true}, "leftValue": "={{ $json.metadata.id }}", "rightValue": ""}]}, "renameOutput": true}]}, "options": {}}, "typeVersion": 3.2}, {"id": "6c5d8f6a-569e-4f1e-99a6-07ec492575ff", "name": "When chat message received", "type": "@n8n/n8n-nodes-langchain.chatTrigger", "position": [660, 2060], "webhookId": "4e762668-c19f-40ec-83bf-302bb9fc6527", "parameters": {"mode": "webhook", "public": true, "options": {}}, "typeVersion": 1.1}, {"id": "9a2f17ba-902f-4528-9eef-f8c0e4ddf516", "name": "Supabase - Retrieve documents from chatinput", "type": "@n8n/n8n-nodes-langchain.vectorStoreSupabase", "position": [1380, 2060], "parameters": {"mode": "load", "prompt": "={{ $json.chatInput }}", "options": {}, "tableName": {"__rl": true, "mode": "list", "value": "documents", "cachedResultName": "documents"}}, "typeVersion": 1}, {"id": "43607f23-d33f-4aca-b478-f20ba8c218cf", "name": "AI Agent", "type": "@n8n/n8n-nodes-langchain.agent", "position": [2780, 2060], "parameters": {"text": "=Visitor's question : {{ $json.chatInput }}\nDocuments found: {{ $json.documents }}", "agent": "conversationalAgent", "options": {"systemMessage": "You are an assistant tasked with answering questions from visitors to the website {{your_website_url}}.\n\nInput:\nVisitor's question: The question posed by the visitor.\nDocuments found: A selection of documents from the vector database that match the visitor's question. These documents are accompanied by the following metadata:\nurl: The URL of the page or blog post found.\ncontent_type: The type of content (e.g., page or blog article).\npublication_date: The publication date of the document.\nmodification_date: The last modification date of the document.\nObjective:\nProvide a helpful answer using the relevant information from the documents found.\nIMPORTANT : You must always include all metadata (url, content_type, publication_date, and modification_date) directly in the main answer to the visitor to indicate the source of the information. These should not be separated from the main answer, and must be naturally integrated into the response.\nIf multiple documents are used in your response, mention each one with its respective metadata.\nIf no relevant documents are found, or if the documents are insufficient, clearly indicate this in your response.\nImportant: Respond in the language used by the visitor who asked the question.\nExample of forced metadata integration:\n\"The cost of a home charging station for an electric vehicle varies depending on several factors. According to [title of the page](https://example.com/charging-point-price), published on April 8, 2021, and updated on July 24, 2022, the price for a 7kW station is \u20ac777.57 including VAT. This page provides further details about the price range and installation considerations.\""}, "promptType": "define"}, "typeVersion": 1.6}, {"id": "cd4107cb-e521-4c1e-88e2-3417a12fd585", "name": "Supabase Vector Store", "type": "@n8n/n8n-nodes-langchain.vectorStoreSupabase", "position": [2940, 900], "parameters": {"mode": "insert", "options": {"queryName": "match_documents"}, "tableName": {"__rl": true, "mode": "list", "value": "documents", "cachedResultName": "documents"}}, "typeVersion": 1}], "active": false, "pinData": {}, "settings": {"executionOrder": "v1"}, "versionId": "fe2a25f4-04b3-462c-97cd-a173b4a0631b", "connections": {"Switch": {"main": [[{"node": "Supabase - Delete row if documents exists", "type": "main", "index": 0}], [{"node": "Set fields4", "type": "main", "index": 0}]]}, "AI Agent": {"main": [[{"node": "Respond to Webhook", "type": "main", "index": 0}]]}, "Postgres": {"main": [[{"node": "Wordpress - Get posts modified after last workflow execution", "type": "main", "index": 0}, {"node": "Wordpress - Get posts modified after last workflow execution1", "type": "main", "index": 0}]]}, "Aggregate": {"main": [[{"node": "Supabase - Store workflow execution", "type": "main", "index": 0}]]}, "Markdown1": {"main": [[{"node": "Store documents on Supabase", "type": "main", "index": 0}]]}, "Aggregate1": {"main": [[{"node": "Store workflow execution id and timestamptz", "type": "main", "index": 0}]]}, "Aggregate2": {"main": [[{"node": "Set fields3", "type": "main", "index": 0}]]}, "Set fields": {"main": [[{"node": "AI Agent", "type": "main", "index": 0}]]}, "Set fields1": {"main": [[{"node": "Filter - Only published & unprotected content", "type": "main", "index": 0}]]}, "Set fields2": {"main": [[{"node": "Filter - Only published and unprotected content", "type": "main", "index": 0}]]}, "Set fields3": {"main": [[{"node": "Loop Over Items", "type": "main", "index": 0}]]}, "Set fields4": {"main": [[{"node": "Loop Over Items", "type": "main", "index": 0}]]}, "Token Splitter": {"ai_textSplitter": [[{"node": "Default Data Loader", "type": "ai_textSplitter", "index": 0}]]}, "Loop Over Items": {"main": [[{"node": "Markdown1", "type": "main", "index": 0}], [{"node": "Postgres - Filter on existing documents", "type": "main", "index": 0}]]}, "Token Splitter1": {"ai_textSplitter": [[{"node": "Default Data Loader1", "type": "ai_textSplitter", "index": 0}]]}, "Every 30 seconds": {"main": [[{"node": "Postgres", "type": "main", "index": 0}]]}, "HTML To Markdown": {"main": [[{"node": "Supabase Vector Store", "type": "main", "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}]]}, "Embeddings OpenAI1": {"ai_embedding": [[{"node": "Supabase - Retrieve documents from chatinput", "type": "ai_embedding", "index": 0}]]}, "Embeddings OpenAI2": {"ai_embedding": [[{"node": "Store documents on Supabase", "type": "ai_embedding", "index": 0}]]}, "Aggregate documents": {"main": [[{"node": "Set fields", "type": "main", "index": 0}]]}, "Default Data Loader": {"ai_document": [[{"node": "Supabase Vector Store", "type": "ai_document", "index": 0}]]}, "Default Data Loader1": {"ai_document": [[{"node": "Store documents on Supabase", "type": "ai_document", "index": 0}]]}, "Postgres Chat Memory": {"ai_memory": [[{"node": "AI Agent", "type": "ai_memory", "index": 0}]]}, "Supabase Vector Store": {"main": [[{"node": "Aggregate", "type": "main", "index": 0}]]}, "Wordpress - Get all pages": {"main": [[{"node": "Merge Wordpress Posts and Pages", "type": "main", "index": 1}]]}, "Wordpress - Get all posts": {"main": [[{"node": "Merge Wordpress Posts and Pages", "type": "main", "index": 0}]]}, "When chat message received": {"main": [[{"node": "Supabase - Retrieve documents from chatinput", "type": "main", "index": 0}]]}, "Store documents on Supabase": {"main": [[{"node": "Aggregate1", "type": "main", "index": 0}]]}, "Merge Wordpress Posts and Pages": {"main": [[{"node": "Set fields1", "type": "main", "index": 0}]]}, "Postgres - Create documents table": {"main": [[{"node": "Postgres - Create workflow execution history table", "type": "main", "index": 0}]]}, "When clicking \u2018Test workflow\u2019": {"main": [[{"node": "Wordpress - Get all posts", "type": "main", "index": 0}, {"node": "Wordpress - Get all pages", "type": "main", "index": 0}]]}, "Postgres - Filter on existing documents": {"main": [[{"node": "Switch", "type": "main", "index": 0}]]}, "Merge retrieved WordPress posts and pages": {"main": [[{"node": "Set fields2", "type": "main", "index": 0}]]}, "Supabase - Delete row if documents exists": {"main": [[{"node": "Aggregate2", "type": "main", "index": 0}]]}, "Supabase - Retrieve documents from chatinput": {"main": [[{"node": "Aggregate documents", "type": "main", "index": 0}]]}, "Filter - Only published & unprotected content": {"main": [[{"node": "HTML To Markdown", "type": "main", "index": 0}]]}, "Filter - Only published and unprotected content": {"main": [[{"node": "Loop Over Items", "type": "main", "index": 0}]]}, "Wordpress - Get posts modified after last workflow execution": {"main": [[{"node": "Merge retrieved WordPress posts and pages", "type": "main", "index": 0}]]}, "Wordpress - Get posts modified after last workflow execution1": {"main": [[{"node": "Merge retrieved WordPress posts and pages", "type": "main", "index": 1}]]}}}

How to Import This Workflow

  1. 1Copy the workflow JSON above using the Copy Workflow JSON button.
  2. 2Open your n8n instance and go to Workflows.
  3. 3Click Import from JSON and paste the copied workflow.

Don't have an n8n instance? Start your free trial at n8nautomation.cloud

Related Templates

Summarize emails with A.I. then send to messenger

Automatically summarize incoming emails using artificial intelligence and then instantly share those summaries to your preferred messenger application with this efficient n8n workflow. This automation connects your IMAP email account to an AI summarization service via an HTTP Request, extracting key information from your messages. The summarized content is then sent to a messenger platform, also through an HTTP Request, ensuring important updates reach your team or personal devices without delay. Businesses can leverage this to quickly disseminate critical client communications, project updates, or urgent alerts from various email accounts directly to team chat channels, saving valuable time spent sifting through lengthy emails. This workflow significantly reduces manual effort in information processing and dissemination, allowing for faster decision-making and improved communication efficiency across your organization.

7 nodes

Send specific PDF attachments from Gmail to Google Drive using OpenAI

Automatically extract and categorize specific PDF attachments from incoming Gmail emails and upload them to designated Google Drive folders based on their content, leveraging OpenAI's advanced text analysis capabilities. This marketing workflow begins with the Gmail trigger "On email received," which then checks if the email "Has attachments?" If attachments are present, the workflow iterates through them, identifying if each is a PDF. For each identified PDF, the "Read PDF" node extracts its textual content. This content is then evaluated by the "Is text within token limit?" node to ensure it's suitable for OpenAI processing. If within limits, the "OpenAI matches PDF textual content" node analyzes the text against predefined criteria. Based on whether the PDF is "matched" by OpenAI, the "Upload file to folder" node in Google Drive stores the relevant PDFs, while unmatched or non-PDF attachments are handled accordingly. This workflow is ideal for marketing teams needing to automatically sort client contracts, campaign reports, or specific vendor invoices received via email, ensuring critical documents are filed correctly without manual intervention. It significantly reduces the time spent on document organization and improves data accessibility, allowing teams to focus on strategic marketing initiatives rather than administrative tasks.

18 nodes

Effortless Email Management with AI

Automate your email management by intelligently processing incoming messages, summarizing their content, and drafting personalized replies with AI. This powerful workflow connects your Gmail inbox via an IMAP trigger to a Text Classifier that categorizes emails, then uses OpenAI for summarization and drafting responses, and Qdrant for vector storage of relevant documents from Google Drive. It’s ideal for marketing teams, customer support, or busy professionals who need to efficiently handle high volumes of correspondence, saving significant time on reading, understanding, and responding to emails by leveraging AI to streamline communication and ensure timely, relevant replies. The system also allows for reviewing AI-generated drafts before sending, ensuring quality control while drastically reducing manual effort.

31 nodes

Ready to automate with n8n?

Get affordable managed n8n hosting with 24/7 support.