{ "id": "QiEoooyNMeJcvKKg", "name": "Automating WhatsApp replies in Go High Level with Redis and Anthropic", "tags": [ { "id": "PnRGSMhEDCtkgPVA", "name": "OJAUUMDev", "createdAt": "2025-07-21T20:12:28.081Z", "updatedAt": "2025-07-21T20:12:28.081Z" } ], "nodes": [ { "id": "6290c10a-80ac-46b8-9d49-dcf5dff98125", "name": "Webhook", "type": "n8n-nodes-base.webhook", "position": [ -1380, 320 ], "parameters": { "path": "fce352e8-8edf-447c-9b54-c2c2b6649ccc", "options": {}, "httpMethod": "POST" }, "typeVersion": 2 }, { "id": "4f19d6b1-829c-4824-8e75-e0781caa59f3", "name": "Switch1", "type": "n8n-nodes-base.switch", "position": [ -1220, 320 ], "parameters": { "rules": { "values": [ { "conditions": { "options": { "version": 1, "leftValue": "", "caseSensitive": true, "typeValidation": "strict" }, "combinator": "and", "conditions": [ { "id": "436fa76a-f948-4de3-86a1-9ed6e568430f", "operator": { "type": "string", "operation": "notContains" }, "leftValue": "={{ $json.body.message.body }}", "rightValue": "type message: audio" } ] } }, { "conditions": { "options": { "version": 1, "leftValue": "", "caseSensitive": true, "typeValidation": "strict" }, "combinator": "and", "conditions": [ { "id": "a4574272-ebee-40a4-a1fb-9139e5f6112e", "operator": { "type": "string", "operation": "contains" }, "leftValue": "={{ $json.body.message.body }}", "rightValue": "type message: audio" } ] } } ] }, "options": {} }, "typeVersion": 3 }, { "id": "2f44dc60-d93f-4ee0-8c5a-e481541d6486", "name": "AI Agent", "type": "@n8n/n8n-nodes-langchain.agent", "position": [ 460, 160 ], "parameters": { "text": "={{ $('messages').first().json.menssage }}", "options": { "systemMessage": "=# Summary\n\nYou are the customer service assistant for **GlobalPoint Services**. Your task is to answer user questions **only using the information provided by the `ClientInfo` tool**, which contains accurate data about services, branches, locations, schedules, prices, and requirements.\n\n# Tools\n\n**ClientInfo** \u2192 Use this tool to obtain the exact information. Its content consists of a list of objects under the `\"response\"` field with the following structure:\n\n```json\n{\n \"response\": {\n \"Category\": \"Consulting\",\n \"Service\": \"Business Strategy Session\",\n \"Requirements\": \"Schedule at least 48 hours in advance\",\n \"Price (USD)\": 150\n }\n},\n{\n \"response\": {\n \"Branch\": \"GlobalPoint Downtown Office\",\n \"Location\": \"123 Main Street, Downtown, Springfield, USA\",\n \"Schedule\": \"Mon\u2013Fri: 9 a.m.\u20136 p.m.; Sat: 9 a.m.\u20131 p.m.; Sun: Closed\",\n \"Notes\": \"Walk-ins are welcome during business hours.\"\n }\n}\n```\n\n# Rules\n\n* When a user asks about branches, locations, schedules, services, prices, or requirements, **always use ClientInfo to find exact matches**, regardless of case sensitivity or minor typos.\n* You may accept partial matches or synonyms as long as they clearly refer to the exact service or branch listed in ClientInfo.\n* Always respond using the exact service or branch name **as it appears in ClientInfo**, and clarify:\n `The price for \"[exact name]\" is XX USD.`\n or\n `The branch \"[exact name]\" is located at [exact location]. Schedule: [schedule].`\n* **Never provide information that is not explicitly included in ClientInfo.**\n* If no clear match is found, respond with:\n `Sorry, we do not offer that service or have that branch in our records. Is there anything else I can help you with?`\n* Use a **friendly, clear, simple, and helpful** tone, always in the **same language as the user**.\n* If the question is generic or just a greeting, respond politely and invite them to ask.\n\n# Examples\n\n**1. Input:** Hello\n**Response:** Hi! Welcome to GlobalPoint Services. How can I assist you today?\n\n**2. Input:** I have a question\n**Response:** Sure, tell me more so I can help.\n\n**3. Input:** What services do you offer?\n**Action:** Read all items in ClientInfo.\n**Response:** At GlobalPoint Services, we offer options such as \\[brief list based on Category and Service].\n\n**4. Input:** How much is the Business Strategy Session?\n**Action:** Search \"Business Strategy Session\" in ClientInfo.\n**Response:** The price for the service \"Business Strategy Session\" is 150 USD. Requirements: Schedule at least 48 hours in advance.\n\n**5. Input:** Where is the Downtown Office?\n**Action:** Search \"Downtown Office\" in branches within ClientInfo.\n**Response:** Our branch \"GlobalPoint Downtown Office\" is located at 123 Main Street, Downtown, Springfield, USA. Schedule: Mon\u2013Fri: 9 a.m.\u20136 p.m. and Sat: 9 a.m.\u20131 p.m.\n\n# Final Reminders\n\n* **Never make up data, prices, or addresses.**\n* **Never answer from general knowledge**, only from ClientInfo.\n* **Always use the exact service or branch name when giving prices or requirements.**\n* **Do not give suggestions or recommendations not present in the data.**\n\n" }, "promptType": "define" }, "typeVersion": 1.7 }, { "id": "e60eaf2a-9c68-49bf-b237-d0bbd6bdd1c7", "name": "Anthropic Chat Model", "type": "@n8n/n8n-nodes-langchain.lmChatAnthropic", "position": [ 380, 380 ], "parameters": { "model": { "__rl": true, "mode": "list", "value": "claude-sonnet-4-20250514", "cachedResultName": "Claude Sonnet 4" }, "options": {} }, "typeVersion": 1.3 }, { "id": "a29f6bab-4088-48f2-802a-c59f44d0d3f5", "name": "Update Contact", "type": "n8n-nodes-base.httpRequest", "position": [ 1280, 160 ], "parameters": { "url": "=https://rest.gohighlevel.com/v1/contacts/{{ $('Webhook').item.json.body.contact_id }}", "method": "PUT", "options": {}, "jsonBody": "={\n \"customField\": {\n \"{{ $json.customFields.find(field => field.name === 'IA_answer').id }}\": \"{{ $('Sanitize and Format Message Output').item.json.output }}\"\n }\n}", "sendBody": true, "sendHeaders": true, "specifyBody": "json", "authentication": "genericCredentialType", "genericAuthType": "httpBearerAuth", "headerParameters": { "parameters": [ {} ] } }, "credentials": { "httpBearerAuth": { "name": "" } }, "typeVersion": 4.2 }, { "id": "ba1cfe67-f7ce-4f32-82a9-99557f4b6495", "name": "GHL Custom fields", "type": "n8n-nodes-base.httpRequest", "position": [ 1080, 160 ], "parameters": { "url": "https://rest.gohighlevel.com/v1/custom-fields/", "options": {}, "sendHeaders": true, "authentication": "genericCredentialType", "genericAuthType": "httpBearerAuth", "headerParameters": { "parameters": [ { "name": "Content-Type", "value": "application/json" } ] } }, "credentials": { "httpBearerAuth": { "name": "" } }, "typeVersion": 4.2 }, { "id": "5ab4f7b8-ccf3-440b-a9ea-54e6d5e595f7", "name": "Wait", "type": "n8n-nodes-base.wait", "position": [ -500, 220 ], "parameters": { "amount": 15 }, "typeVersion": 1.1 }, { "id": "b081b133-b582-447e-bf52-8647d4c18e24", "name": "If", "type": "n8n-nodes-base.if", "position": [ -180, 220 ], "parameters": { "options": {}, "conditions": { "options": { "version": 2, "leftValue": "", "caseSensitive": true, "typeValidation": "strict" }, "combinator": "and", "conditions": [ { "id": "41b0355b-7204-4e25-b74a-7bf70dea8a9c", "operator": { "name": "filter.operator.equals", "type": "string", "operation": "equals" }, "leftValue": "={{ $json.messages.last() }}", "rightValue": "={{ $('Sanitize and Format Message Body').item.json.text }}" } ] } }, "typeVersion": 2.2 }, { "id": "1f6b0707-cd9d-4d93-ba39-757348beceb1", "name": "No Operation, do nothing", "type": "n8n-nodes-base.noOp", "position": [ -20, 380 ], "parameters": {}, "typeVersion": 1 }, { "id": "86cb66f8-0d79-40ee-88c6-48ac440a08cf", "name": "Simple Memory", "type": "@n8n/n8n-nodes-langchain.memoryBufferWindow", "position": [ 520, 380 ], "parameters": { "sessionKey": "={{ $('Webhook').item.json.body.phone }}", "sessionIdType": "customKey" }, "typeVersion": 1.3 }, { "id": "27d0db94-0317-423f-8b58-8178211f8dfb", "name": "Save message", "type": "n8n-nodes-base.redis", "position": [ -660, 220 ], "parameters": { "list": "={{ $('Webhook').item.json.body.contact_id }}", "tail": true, "operation": "push", "messageData": "={{ $json.text }}" }, "typeVersion": 1 }, { "id": "6c8b6cb0-7cd7-4e3d-90c4-4e117f3449eb", "name": "Get messages", "type": "n8n-nodes-base.redis", "position": [ -340, 220 ], "parameters": { "key": "={{ $('Webhook').item.json.body.contact_id }}", "options": {}, "operation": "get", "propertyName": "mensajes" }, "typeVersion": 1 }, { "id": "877a16b7-b461-439f-8501-9a4c418dc7d0", "name": "Sanitize and Format Message Body", "type": "n8n-nodes-base.code", "position": [ -940, 220 ], "parameters": { "jsCode": "return [{\n text: $input.first().json.body.message.body\n .replace(/\\\\+\"/g, '\"') // Fix incorrectly escaped double quotes\n .replace(/\\n/g, \"\\\\n\") // Properly escape line breaks\n .replace(/\\r/g, \"\\\\r\") // Escape carriage returns\n .replace(/\\t/g, \"\\\\t\") // Escape tab characters\n .trim() // Remove unnecessary whitespace\n}];" }, "typeVersion": 2 }, { "id": "8c8a0ce0-4fc3-4979-8f1a-331a0861bfe4", "name": "Delete messages", "type": "n8n-nodes-base.redis", "position": [ 120, 160 ], "parameters": { "key": "={{ $('Webhook').item.json.body.contact_id }}", "operation": "delete" }, "typeVersion": 1 }, { "id": "55e65c69-54ec-4df1-8197-0dd0b4f6f416", "name": "messages", "type": "n8n-nodes-base.set", "position": [ -20, 160 ], "parameters": { "options": {}, "assignments": { "assignments": [ { "id": "7bb3ef98-b766-4ad7-8343-b2885bd70225", "name": "message", "type": "string", "value": "={{ $node[\"Get messages\"].json[\"messages\"].join(\"\\n\") }}" }, { "id": "62dfc214-fb40-4a59-81a4-b01c6c0860af", "name": "id", "type": "string", "value": "={{ $('Webhook').item.json.body.contact_id }}" } ] } }, "typeVersion": 3.4 }, { "id": "e65eca71-742f-4301-bbab-5130fe7e4d34", "name": "When Executed by Another Workflow", "type": "n8n-nodes-base.executeWorkflowTrigger", "position": [ -600, 1140 ], "parameters": { "inputSource": "passthrough" }, "typeVersion": 1.1 }, { "id": "22b16984-474c-488a-b012-17d6cc8f26c9", "name": "Download file", "type": "n8n-nodes-base.googleDrive", "position": [ -440, 1140 ], "parameters": { "fileId": { "__rl": true, "mode": "list", "value": "1Im72SvVQhzjLk_rVUidGK0du4s2zMZ3H", "cachedResultUrl": "https://docs.google.com/spreadsheets/d/1Im72SvVQhzjLk_rVUidGK0du4s2zMZ3H/edit?usp=drivesdk&ouid=107080324345235325088&rtpof=true&sd=true", "cachedResultName": "Examenes de Laboratorio Prueba Doctor Premier.xlsx" }, "options": {}, "operation": "download" }, "credentials": { "googleDriveOAuth2Api": { "name": "" } }, "typeVersion": 3 }, { "id": "aabc26f2-36a2-469c-8c60-5a7b35030c16", "name": "Success", "type": "n8n-nodes-base.set", "position": [ 80, 1140 ], "parameters": { "options": {}, "assignments": { "assignments": [ { "id": "39c2f302-03be-4464-a17a-d7cc481d6d44", "name": "=response", "type": "object", "value": "={{ $json }}" } ] } }, "typeVersion": 3.4 }, { "id": "ce02d432-ecca-44ff-ac7e-2f3d12f09185", "name": "Merge", "type": "n8n-nodes-base.merge", "position": [ -60, 1140 ], "parameters": {}, "typeVersion": 3.2 }, { "id": "e9836dfc-8a86-41be-ac81-5e706ce38a23", "name": "Call n8n Workflow Tool", "type": "@n8n/n8n-nodes-langchain.toolWorkflow", "position": [ 680, 380 ], "parameters": { "workflowId": { "__rl": true, "mode": "list", "value": "yVUPXPNCNlo5HkJs", "cachedResultName": "My Sub-Workflow 1" }, "workflowInputs": { "value": {}, "schema": [], "mappingMode": "defineBelow", "matchingColumns": [], "attemptToConvertTypes": false, "convertFieldsToString": false } }, "typeVersion": 2.2 }, { "id": "7a881605-ce1d-4ed4-9c61-36533b759286", "name": "Extract Tests", "type": "n8n-nodes-base.extractFromFile", "position": [ -240, 1060 ], "parameters": { "options": { "sheetName": "Tests" }, "operation": "xlsx" }, "typeVersion": 1 }, { "id": "44d5d6c6-add6-424c-ad6c-0108435a1c1b", "name": "Extract Sites", "type": "n8n-nodes-base.extractFromFile", "position": [ -240, 1240 ], "parameters": { "options": { "sheetName": "Sites" }, "operation": "xlsx", "binaryPropertyName": "=data" }, "typeVersion": 1 }, { "id": "b8d3874e-ee58-4da7-bcf5-3b6b67071411", "name": "Sticky Note", "type": "n8n-nodes-base.stickyNote", "position": [ -660, 580 ], "parameters": { "color": 5, "width": 900, "height": 820, "content": "# Google Drive Excel Merge Workflow\n\nThis is an **optional example** that should be created in a **separate workflow**. It demonstrates how to extract client information from an Excel file with two sheets, `tests` and `sites`, merge the data, and return it to an LLM.\n\n## 1. Setup and Download\n* Create Google OAuth 2.0 credentials in **Google Cloud Console** and enable the **Google Drive API**.\n* Authenticate and download the specified Excel file from Google Drive.\n---\n## 2. Extract Data from Excel\n* Read the Excel file.\n* Extract data from the **tests** sheet.\n* Extract data from the **sites** sheet.\n---\n## 3. Merge the Data\n* Combine the extracted data from both sheets into one dataset.\n* Ensure formatting and structure are consistent.\n* Return the merged dataset as the response to the LLM request.\n" }, "typeVersion": 1 }, { "id": "62d3abdc-7eac-4552-a064-23097d6ea7d6", "name": "Sticky Note1", "type": "n8n-nodes-base.stickyNote", "position": [ -1400, -300 ], "parameters": { "width": 680, "height": 860, "content": "# GHL to n8n SMS Reply Bridge\n\n## 1. Requirements and Setup\n\n* Requires the **GoHighLevel sub-account API** (API key/token) where the chatbot will run.\n* In GoHighLevel, create an **Automation**: trigger on **Customer Replied (SMS)** and send the payload to your **n8n Webhook** URL.\n* Use the **Wazzap** plugin in GHL to receive WhatsApp replies into the same flow.\n* Cost-Saving Trick: Create a custom field in GHL, for example **IA_answer** and update with your bot\u2019s response instead of triggering extra paid automations on every inbound message.\n---\n\n## 2. Webhook and Filter\n\n* Receive the HTTP POST in **n8n Webhook**.\n* Add a filter/switch:\n * If the message is **audio** (Wazzap/GHL does not provide a downloadable file), **do not process**. Send a personalized fallback reply asking for text.\n * If the message is **text**, run a **Function/Code** node to **sanitize**:\n * Fix incorrectly escaped quotes.\n * Escape line breaks, carriage returns, and tabs.\n * Trim whitespace and remove invalid fields.\n\n" }, "typeVersion": 1 }, { "id": "20e82e23-0434-495e-9676-0cf6d91f1c5a", "name": "Update Contact No audio", "type": "n8n-nodes-base.httpRequest", "position": [ -860, 400 ], "parameters": { "url": "=https://rest.gohighlevel.com/v1/contacts/{{ $('Webhook').item.json.body.contact_id }}", "method": "PUT", "options": {}, "jsonBody": "={\n \"customField\": {\n \"{{ $json.customFields.find(field => field.name === 'IA_answer').id }}\": \"Sorry, I can\u2019t listen to voice notes at the moment. Could you please write your question or concern? Thank you very much.\"\n }\n}", "sendBody": true, "sendHeaders": true, "specifyBody": "json", "authentication": "genericCredentialType", "genericAuthType": "httpBearerAuth" }, "credentials": { "httpBearerAuth": { "name": "" } }, "typeVersion": 4.2 }, { "id": "ad9f2916-0f90-4c4a-85eb-6b7dfc705fa8", "name": "Get GHL Custom fields", "type": "n8n-nodes-base.httpRequest", "position": [ -1060, 400 ], "parameters": { "url": "https://rest.gohighlevel.com/v1/custom-fields/", "options": {}, "sendHeaders": true, "authentication": "genericCredentialType", "genericAuthType": "httpBearerAuth", "headerParameters": { "parameters": [ { "name": "Content-Type", "value": "application/json" } ] } }, "credentials": { "httpBearerAuth": { "name": "" } }, "typeVersion": 4.2 }, { "id": "1cd6d85e-62e7-41d4-a4a8-5e6ded3140de", "name": "Sticky Note2", "type": "n8n-nodes-base.stickyNote", "position": [ -660, -300 ], "parameters": { "color": 6, "width": 900, "height": 860, "content": "# Redis Message Buffer Workflow\n\n## 1. Purpose\n* Batch multiple user messages sent within a short window and pass a single merged text to the next step.\n---\n## 2. Setup\n* Configure a Redis database and set credentials/values in the Redis nodes (host, port, db index, password).\n* Use one Redis list per contact, keyed by the `contact_id` from the Webhook.\n---\n## 3. Flow\n* **Save message**: Push the sanitized text to the Redis list for `contact_id`.\n* **Wait**: Pause for a few seconds (e.g., 15) so the user can finish typing.\n* **Get messages**: Read all items from the Redis list into `messages`.\n* **If**: Compare `messages.last()` with the latest sanitized text; if equal proceed, if not do nothing.\n* **messages**: Join all collected items with newline into `message` and keep `id = contact_id`.\n* **Delete messages**: Delete the Redis key for that `contact_id` to clear the buffer.\n" }, "typeVersion": 1 }, { "id": "6949cf34-0086-4493-b281-2d8d200c7fcb", "name": "Sanitize and Format Message Output", "type": "n8n-nodes-base.code", "position": [ 800, 160 ], "parameters": { "jsCode": "return [{\n output: $input.first().json.output\n .replace(/^\"{|}\"$/g, \"\") // Removes double quotes at the start and end if present\n .replace(/^{\"response\":\\s*\"/, '') // Removes '{\"response\":\"' at the beginning if present\n .replace(/\"}$/, '') // Removes '\"}' at the end if present\n .replace(/\\\\\"/g, '\"') // Converts escaped quotes \\\" into normal quotes \"\n .replace(/\\n/g, \"\\\\n\") // Escapes newlines correctly\n .replace(/\\r/g, \"\\\\r\") // Escapes carriage returns\n .replace(/\\t/g, \"\\\\t\") // Escapes tabs\n .replace(/^\"+|\"+$/g, \"\") // \ud83d\udd25 New line: Removes extra double quotes at the edges\n .replace(/\"/g, '\\\\\"') // Escapes internal double quotes for JSON\n .trim() // Removes unnecessary whitespace\n}];" }, "typeVersion": 2 }, { "id": "99f56ece-0c95-4148-938a-51e9d32d047f", "name": "Sticky Note3", "type": "n8n-nodes-base.stickyNote", "position": [ 340, -300 ], "parameters": { "color": 3, "width": 660, "height": 860, "content": "# ClientInfo LLM Response Workflow\n\n## 1. AI Agent with ClientInfo Tool\n* Answers user questions **only** with data from **ClientInfo** (sub-workflow below), calling it through **Call n8n Workflow Tool**.\n * Uses **Anthropic (Claude Sonnet 4)** in this example but can work with any chat model by setting the API key.\n * The system prompt enforces answering only from ClientInfo, and can be tuned for your client\u2019s tone, style, or specificity based on available data.\n * Keeps short conversational context with **Simple Memory** using the user\u2019s phone as the session key.\n---\n## 2. Sanitize Output\n* After generating the response, a **Function/Code** node cleans it by fixing incorrectly escaped quotes, escaping line breaks, carriage returns, and tabs, trimming whitespace, and removing invalid fields before sending it downstream.\n" }, "typeVersion": 1 }, { "id": "3608f575-00cf-42a8-acd8-44771e584288", "name": "Sticky Note4", "type": "n8n-nodes-base.stickyNote", "position": [ 1020, -300 ], "parameters": { "width": 460, "height": 620, "content": "# GHL Update Contact Workflow\n\n## 1. Get Custom Fields\n * Makes a **GET** request to the GoHighLevel API to retrieve all custom fields in the sub-account.\n * Finds the **IA_answer** field ID to be used in the update request.\n---\n## 2. Update Contact\n * Sends a **PUT** request to the GoHighLevel API to update the contact\u2019s **IA_answer** custom field with the sanitized AI response.\n * In GHL, create an **Automation** that triggers when **IA_answer** is updated and sends the value by SMS to the contact.\n" }, "typeVersion": 1 }, { "id": "996bfe66-4b59-4702-aa59-55da2c5282e3", "name": "Sticky Note5", "type": "n8n-nodes-base.stickyNote", "position": [ -2600, -300 ], "parameters": { "color": 3, "width": 1120, "height": 1040, "content": "# Automating WhatsApp replies in Go High Level with Redis and Anthropic\n## Description\n* Integrates GHL + Wazzap with Redis and an AI Agent using ClientInfo to process messages, generate accurate replies, and send them via a custom field trigger.\n---\n## Who\u2019s it for\n* This workflow is for businesses using **GoHighLevel (GHL)**, including the **Wazzap** plugin for WhatsApp, who want to automate inbound SMS/WhatsApp replies with AI. It\u2019s ideal for teams that need accurate, data-driven responses from a predefined **ClientInfo** source and want to send them back to customers without paying for extra inbound automations.\n---\n## How it works / What it does\n1. **Receive message** in n8n via Webhook from GHL (**Customer Replied (SMS)** automation). WhatsApp messages arrive the same way using the **Wazzap** plugin.\n2. **Filter message type**:\n * If audio \u2192 skip processing and send fallback asking for text.\n * If text \u2192 sanitize by fixing escaped quotes, escaping line breaks/carriage returns/tabs, and removing invalid fields.\n3. **Buffer messages** in Redis to group multiple messages sent in a short window.\n4. **Run AI Agent** using the **ClientInfo** tool to answer only with accurate service/branch data.\n5. **Sanitize AI output** before sending back.\n6. **Update GHL contact** custom field (**IA_answer**) with the AI\u2019s response.\n7. **Send SMS reply** automatically via GHL\u2019s outbound automation triggered by the updated custom field.\n---\n## How to set up\n1. In **GHL**, create:\n * **Inbound automation**: Trigger on *Customer Replied (SMS)* \u2192 Send to your n8n Webhook.\n * **Outbound automation**: Trigger when **IA_answer** is updated \u2192 Send SMS to the contact.\n * Create a custom field named **IA_answer**.\n2. Connect **Wazzap** in GHL to handle WhatsApp messages.\n3. Configure **Redis** in n8n (host, port, DB index, password).\n4. Add your **AI model** credentials (Anthropic, OpenAI, etc.) in n8n.\n5. (Optional) Set up the **Google Drive Excel Merge** sub-workflow to enrich ClientInfo with external data.\n---\n## Requirements\n* **GoHighLevel sub-account API key**.\n* **Anthropic (Claude)** API key or another supported LLM provider.\n* **Redis database** for temporary message storage.\n* **GHL automations**: one for inbound messages to n8n, one for outbound replies when **IA\\_answer** is updated.\n* **GHL custom field**: **IA\\_answer** to store and trigger replies.\n* **Wazzap plugin** in GHL for WhatsApp message handling.\n---\n## How to customize the workflow\n* Add more context or business-specific data to the **AI Agent prompt** so replies match your brand tone and policies.\n* Expand the **ClientInfo** dataset with additional services, branches, or product details.\n* Adjust the **Redis wait time** to control how long the workflow buffers messages before replying." }, "typeVersion": 1 } ], "active": false, "settings": { "executionOrder": "v1" }, "versionId": "7ff7549a-738b-4be0-8c59-bdd9f605edfa", "connections": { "If": { "main": [ [ { "node": "messages", "type": "main", "index": 0 } ], [ { "node": "No Operation, do nothing", "type": "main", "index": 0 } ] ] }, "Wait": { "main": [ [ { "node": "Get messages", "type": "main", "index": 0 } ] ] }, "Merge": { "main": [ [ { "node": "Success", "type": "main", "index": 0 } ] ] }, "Switch1": { "main": [ [ { "node": "Sanitize and Format Message Body", "type": "main", "index": 0 } ], [ { "node": "Get GHL Custom fields", "type": "main", "index": 0 } ] ] }, "Webhook": { "main": [ [ { "node": "Switch1", "type": "main", "index": 0 } ] ] }, "AI Agent": { "main": [ [ { "node": "Sanitize and Format Message Output", "type": "main", "index": 0 } ] ] }, "messages": { "main": [ [ { "node": "Delete messages", "type": "main", "index": 0 } ] ] }, "Get messages": { "main": [ [ { "node": "If", "type": "main", "index": 0 } ] ] }, "Save message": { "main": [ [ { "node": "Wait", "type": "main", "index": 0 } ] ] }, "Download file": { "main": [ [ { "node": "Extract Tests", "type": "main", "index": 0 }, { "node": "Extract Sites", "type": "main", "index": 0 } ] ] }, "Extract Sites": { "main": [ [ { "node": "Merge", "type": "main", "index": 1 } ] ] }, "Extract Tests": { "main": [ [ { "node": "Merge", "type": "main", "index": 0 } ] ] }, "Simple Memory": { "ai_memory": [ [ { "node": "AI Agent", "type": "ai_memory", "index": 0 } ] ] }, "Delete messages": { "main": [ [ { "node": "AI Agent", "type": "main", "index": 0 } ] ] }, "GHL Custom fields": { "main": [ [ { "node": "Update Contact", "type": "main", "index": 0 } ] ] }, "Anthropic Chat Model": { "ai_languageModel": [ [ { "node": "AI Agent", "type": "ai_languageModel", "index": 0 } ] ] }, "Get GHL Custom fields": { "main": [ [ { "node": "Update Contact No audio", "type": "main", "index": 0 } ] ] }, "Call n8n Workflow Tool": { "ai_tool": [ [ { "node": "AI Agent", "type": "ai_tool", "index": 0 } ] ] }, "Sanitize and Format Message Body": { "main": [ [ { "node": "Save message", "type": "main", "index": 0 } ] ] }, "When Executed by Another Workflow": { "main": [ [ { "node": "Download file", "type": "main", "index": 0 } ] ] }, "Sanitize and Format Message Output": { "main": [ [ { "node": "GHL Custom fields", "type": "main", "index": 0 } ] ] } } }