January 22, 2026

Google Sheets + Google Drive, support replies stay consistent

Lisa Granqvist Partner Workflow Automation Expert

Get a free AI assessment → ⬇️ Use template

Your support answers are probably “right”… but not consistent. One teammate replies from memory, another searches old emails, and someone else copy-pastes a half-updated snippet from a doc.

This Sheets Drive automation hits support leads first, honestly. Ops managers feel it when reporting gets messy. And agency owners notice when clients complain about mixed messages. The outcome: faster, more uniform replies plus a clean log of every conversation for review.

You’ll set up an n8n workflow that pulls context from Google Sheets and Google Drive, uses AI to draft grounded replies, and stores every interaction for weekly reporting.

How This Automation Works

Here’s the complete workflow you’ll be setting up:

n8n Workflow Template: Google Sheets + Google Drive, support replies stay consistent

Click to explore

flowchart LR

    subgraph sg0["When chat message received Flow"]
        direction LR
        n0@{ icon: "mdi:memory", form: "rounded", label: "Short-Term Memory", pos: "b", h: 48 }
        n1@{ icon: "mdi:robot", form: "rounded", label: "AI Agent (Chat Composer)", pos: "b", h: 48 }
        n2@{ icon: "mdi:robot", form: "rounded", label: "Output Parser (JSON Enforcem..", pos: "b", h: 48 }
        n3@{ icon: "mdi:brain", form: "rounded", label: "OpenAI Chat Model", pos: "b", h: 48 }
        n4@{ icon: "mdi:play-circle", form: "rounded", label: "When chat message received", pos: "b", h: 48 }
        n12@{ icon: "mdi:database", form: "rounded", label: "Get Previous Content from Sh..", pos: "b", h: 48 }
        n13@{ icon: "mdi:database", form: "rounded", label: "Conversation Logging", pos: "b", h: 48 }
        n15@{ icon: "mdi:cube-outline", form: "rounded", label: "Pinecone Vector Store Query ..", pos: "b", h: 48 }
        n16@{ icon: "mdi:vector-polygon", form: "rounded", label: "Embeddings OpenAI Query for ..", pos: "b", h: 48 }
        n21@{ icon: "mdi:swap-vertical", form: "rounded", label: "Format Data For AI Agent ", pos: "b", h: 48 }
        n3 -.-> n1
        n0 -.-> n1
        n1 --> n13
        n21 --> n1
        n4 --> n21
        n12 -.-> n1
        n2 -.-> n1
        n16 -.-> n15
        n15 -.-> n1
    end

    subgraph sg1["Schedule Flow"]
        direction LR
        n10@{ icon: "mdi:play-circle", form: "rounded", label: "Schedule Trigger", pos: "b", h: 48 }
        n11@{ icon: "mdi:cog", form: "rounded", label: "Aggregate", pos: "b", h: 48 }
        n17@{ icon: "mdi:database", form: "rounded", label: "Get Data from Sheet.", pos: "b", h: 48 }
        n18@{ icon: "mdi:cog", form: "rounded", label: "Convert Data  to file", pos: "b", h: 48 }
        n19@{ icon: "mdi:swap-horizontal", form: "rounded", label: "Check File Exist", pos: "b", h: 48 }
        n20@{ icon: "mdi:message-outline", form: "rounded", label: "Send Chat History with attac..", pos: "b", h: 48 }
        n11 --> n18
        n19 --> n20
        n10 --> n17
        n17 --> n11
        n18 --> n19
    end

    subgraph sg2["Google Drive Flow"]
        direction LR
        n5@{ icon: "mdi:play-circle", form: "rounded", label: "Google Drive Trigger", pos: "b", h: 48 }
        n6@{ icon: "mdi:cog", form: "rounded", label: "Download file", pos: "b", h: 48 }
        n7@{ icon: "mdi:robot", form: "rounded", label: "Default Data Loader", pos: "b", h: 48 }
        n8@{ icon: "mdi:robot", form: "rounded", label: "Recursive Character Text Spl..", pos: "b", h: 48 }
        n9@{ icon: "mdi:vector-polygon", form: "rounded", label: "Embeddings OpenAI", pos: "b", h: 48 }
        n14@{ icon: "mdi:cube-outline", form: "rounded", label: "Pinecone Vector Store Insert", pos: "b", h: 48 }
        n6 --> n14
        n9 -.-> n14
        n7 -.-> n14
        n5 --> n6
        n8 -.-> n7
    end

    %% Styling
    classDef trigger fill:#e8f5e9,stroke:#388e3c,stroke-width:2px
    classDef ai fill:#e3f2fd,stroke:#1976d2,stroke-width:2px
    classDef aiModel fill:#e8eaf6,stroke:#3f51b5,stroke-width:2px
    classDef decision fill:#fff8e1,stroke:#f9a825,stroke-width:2px
    classDef database fill:#fce4ec,stroke:#c2185b,stroke-width:2px
    classDef api fill:#fff3e0,stroke:#e65100,stroke-width:2px
    classDef code fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px
    classDef disabled stroke-dasharray: 5 5,opacity: 0.5
    class n4,n10,n5 trigger
    class n1,n2,n7,n8 ai
    class n3 aiModel
    class n0 ai
    class n15,n14 ai
    class n16,n9 ai
    class n19 decision
    class n12,n13,n17 database

Why This Matters: Support Answers Drift Fast

Support information spreads out because it has to. Policies live in Google Drive docs, edge cases are buried in past chats, and “official” answers end up as sticky notes or snippets in someone’s personal file. Then a customer asks a simple question, and your team spends five minutes searching, another five minutes guessing, and then you still ship an answer that’s slightly off. Multiply that by a busy week and you get the real cost: slower response times, repeat tickets, and that quiet anxiety that you’re giving different customers different rules.

It adds up fast. Here’s where it breaks down in the day-to-day.

People answer from “what I remember,” so the same question gets three different replies in the same week.
Searching Drive is slow when you don’t know the doc name, which means customers wait while your team hunts.
Even when you find the right info, copy-pasting leads to missing steps, outdated links, or the wrong tone.
Reporting becomes a chore because the chat history isn’t logged in one place with timestamps and intent.

What You’ll Build: Consistent AI Replies Grounded in Your Sources

This workflow turns your existing knowledge into something your team can actually use in real time. A chat message comes in through n8n’s chat trigger (you can connect it to Telegram, Slack, a website widget, or another channel). The workflow enriches that message by extracting intent and key topic, then retrieves two kinds of context: structured “known answers” from Google Sheets and semantic matches from a Pinecone knowledge base populated from your Google Drive files. With that context in hand, an OpenAI chat model drafts a natural reply that stays anchored to what your business has already written. Finally, the workflow appends the full interaction to Google Sheets, so every question, intent, and timestamp becomes searchable later.

In parallel, it also watches a Google Drive folder for new files. When something new shows up, the workflow loads the document, chunks it into sensible pieces, creates embeddings, and upserts them into Pinecone. That way your “source of truth” grows automatically instead of dying in a forgotten folder.

What You’re Building

What Gets Automated

What You’ll Achieve

Incoming support chats trigger an AI-assisted response workflow instantly.
Google Sheets rows are pulled as structured context for policies, links, and standard answers.
Google Drive documents are indexed into Pinecone automatically when new files appear.
Every conversation is appended to a Google Sheet for auditing and reporting.

Most teams cut “search + compose” from about 10 minutes to about 2 minutes per ticket.
Replies stay consistent because they’re generated from the same approved sources.
You get a searchable history with timestamps and detected intent, ready for reviews.
New docs become usable knowledge without anyone “remembering to update the bot.”
Weekly summaries land in Gmail automatically, so nothing gets ignored.

Expected Results

Say your team handles about 20 support chats a day. Manually, it’s often around 10 minutes per chat to search Drive, check a sheet, and write a careful reply, so that’s roughly 3 hours daily. With this workflow, a message triggers instantly, the AI drafts a response using Sheets + Pinecone context, and logging happens automatically. Your human time becomes review and send, maybe 2 minutes per chat. That’s about 2 hours back most days, plus fewer “can you clarify?” follow-ups.

Before You Start

n8n instance (try n8n Cloud free)
Self-hosting option if you prefer (Hostinger works well)
Google Sheets for structured answers and conversation logs.
Google Drive to store and auto-ingest source documents.
OpenAI API key (get it from your OpenAI dashboard)
Pinecone API credentials (get them from Pinecone console)
Gmail to email weekly summaries automatically.

Skill level: Intermediate. You’ll connect a few accounts, paste IDs (sheet, folder, index), and run a couple of test messages.

Want someone to build this for you? Talk to an automation expert (free 15-minute consultation).

Step by Step

A chat message comes in. The workflow starts from n8n’s chat message trigger, so every inbound question kicks off the same process. You can route messages from a channel like Telegram or Slack into that trigger depending on where support happens.

The message gets cleaned up and understood. n8n prepares the agent inputs, keeps light session memory (so follow-up questions don’t lose context), and parses structured fields like topic and intent. This part matters more than people think because it stops the AI from rambling.

Context is pulled from Sheets and your knowledge base. Google Sheets provides structured “known good” info, while Pinecone handles semantic lookup across your Drive documents. Together, you get an answer that is both consistent and relevant, instead of a generic AI response.

A reply is generated and everything is logged. The OpenAI chat model drafts the response, the orchestrator passes it back to the user, and the conversation is appended to Google Sheets with timestamps and intent. Separately, a weekly schedule aggregates logs and emails a summary via Gmail, with an attached file of history.

You can easily modify the Drive folder being monitored or the logging fields in Sheets based on your needs. See the full implementation guide below for customization options.

Step-by-Step Implementation Guide

Step 1: Configure the Chat and Schedule Triggers

Set up the workflow entry points for chat interactions, weekly exports, and Google Drive file ingestion.

Add and configure Chat Message Trigger to receive incoming chat messages.
Add Weekly Schedule Trigger and set Rule to trigger weekly on day 1 at hour 22.
Configure Drive File Created Trigger with Event set to fileCreated and Trigger On set to specificFolder.
Set Folder To Watch to the target folder ID in Drive File Created Trigger.
Credential Required: Connect your googleDriveOAuth2Api credentials in Drive File Created Trigger.

Step 2: Connect Google Sheets and Build the Weekly Export

Pull weekly sheet data, combine it into a single file, validate the output, and email the attachment.

Configure Retrieve Sheet Data with the target Document ID and Sheet Name.
Credential Required: Connect your googleSheetsOAuth2Api credentials in Retrieve Sheet Data.
In Combine Sheet Records, set Aggregate to aggregateAllItemData.
In Convert Records to File, set Operation to toText and Source Property to data.
In Validate File Type, set the condition Left Value to {{ $binary.data.mimeType }} and Right Value to application/json.
Configure Email History Attachment with Subject set to chat_history_{{ $now.format('DD') }} and Message set to Hi [Name], Please find attached your chat history from our conversation on chat_history_{{ $now.format('DD') }}. Feel free to refer back to this anytime, and let me know if you have any questions! Best regards, [Your Name].
Credential Required: Connect your gmailOAuth2 credentials in Email History Attachment.

⚠️ Common Pitfall: If Validate File Type does not detect application/json, the email will not send. Verify the file output type from Convert Records to File.

Step 3: Prepare Chat Inputs

Normalize and enrich incoming chat messages before sending them to the AI agent.

Connect Chat Message Trigger to Prepare Agent Inputs.
In Prepare Agent Inputs, set intent to Chat.
Set topic to AI Seo Basics and content_id to C001.
Set parameter to the expression {{ $json.parameter }} to pass through any upstream metadata.

Step 4: Set Up AI Orchestration and Tools

Configure the AI agent, memory, tools, and output parsing for structured chat responses.

In Chat Response Orchestrator, set Text to User message: {{ $json.content }} Intent: {{ $json.intent }} Topic: {{ $json.topic || "general" }} Context (from Google Sheets or memory): {{ $json.context || $memory || "No context retrieved" }}.
Keep Has Output Parser enabled in Chat Response Orchestrator and connect Structured JSON Parser with schema { "reply": "string", "context_used": ["string"] }.
Attach Session Memory Buffer to Chat Response Orchestrator with Session Key set to chat-rag-session, Session Id Type set to customKey, and Context Window Length set to 7.
Connect OpenAI Chat Engine as the language model for Chat Response Orchestrator.
Credential Required: Connect your openAiApi credentials in OpenAI Chat Engine.
Attach Fetch Sheet Context and Pinecone Knowledge Lookup as tools to Chat Response Orchestrator, and set Top K to 5 with Tool Description set to Retrieve data from the Pinecone knowledge base and use it to answer user queries about the company, products, or projects in a well-structured, human-like manner.
Credential Required: Connect your googleSheetsOAuth2Api credentials in Fetch Sheet Context and your pineconeApi credentials in Pinecone Knowledge Lookup.
Connect Chat Query Embeddings to Pinecone Knowledge Lookup as the embedding provider.
Credential Required: Connect your openAiApi credentials in Chat Query Embeddings.

Structured AI sub-nodes like Session Memory Buffer and Structured JSON Parser do not require credentials; ensure credentials are set on the parent nodes they attach to, such as OpenAI Chat Engine.

Step 5: Configure Knowledge Ingestion to Pinecone

Ingest new Drive files into Pinecone so the agent can answer questions using the knowledge base.

In Retrieve Drive File, set Operation to download and File ID to {{ $json.id }}.
Credential Required: Connect your googleDriveOAuth2Api credentials in Retrieve Drive File.
In Standard Document Loader, set Data Type to binary, Text Splitting Mode to custom, and set metadata file-name to {{ $('Retrieve Drive File').item.json.name }}.
Configure Recursive Text Chunker with Chunk Size set to 500 and Chunk Overlap set to 200.
Attach OpenAI Embedding Builder with Dimensions set to 512, and connect it to Pinecone Index Upsert.
Credential Required: Connect your openAiApi credentials in OpenAI Embedding Builder and your pineconeApi credentials in Pinecone Index Upsert.
In Pinecone Index Upsert, set Mode to insert and Embedding Batch Size to 1000.

Step 6: Log Responses to Google Sheets

Store structured responses and context usage back to Google Sheets for auditing and reporting.

Connect Chat Response Orchestrator to Append Conversation Log.
In Append Conversation Log, set Operation to appendOrUpdate and select the target Document ID and Sheet Name.
Credential Required: Connect your googleSheetsOAuth2Api credentials in Append Conversation Log.

Step 7: Test & Activate Your Workflow

Run end-to-end tests to confirm all chat, ingestion, and logging flows execute as expected.

Use Execute Workflow to send a test message into Chat Message Trigger and verify a structured response is produced.
Confirm Append Conversation Log writes a new row with the parsed reply and context_used fields.
Manually trigger Weekly Schedule Trigger to verify Email History Attachment sends an email with the exported JSON attachment.
Upload a test file into the watched Drive folder to verify Drive File Created Trigger and Pinecone Index Upsert run successfully.
When all tests pass, toggle the workflow to Active for production use.

🔒

Unlock Full Step-by-Step Guide

Get the complete implementation guide + downloadable template

Troubleshooting Tips

Google Sheets credentials can expire or need specific permissions. If things break, check the connected Google account in n8n Credentials and confirm it can edit the target sheet.
If you’re using Drive ingestion plus embeddings, processing times vary with file size. If Pinecone upserts fail or return empty chunks, increase chunking limits or avoid huge PDFs in the monitored folder.
Default prompts in AI nodes are generic. Add your brand voice early (tone, formatting, what not to promise), or you’ll be editing outputs forever.

Quick Answers

What’s the setup time for this Sheets Drive automation automation?

About an hour if your accounts and IDs are ready.

Is coding required for this support reply consistency?

No. You’ll mostly connect credentials and paste in your Google Sheet ID, Drive folder ID, and Pinecone index name.

Is n8n free to use for this Sheets Drive automation workflow?

Yes. n8n has a free self-hosted option and a free trial on n8n Cloud. Cloud plans start at $20/month for higher volume. You’ll also need to factor in OpenAI API usage (often a few cents per conversation) and Pinecone storage/query costs depending on your document size.

Where can I host n8n to run this automation?

Two options: n8n Cloud (managed, easiest setup) or self-hosting on a VPS. For self-hosting, Hostinger VPS is affordable and handles n8n well. Self-hosting gives you unlimited executions but requires basic server management.

Can I modify this Sheets Drive automation workflow for different use cases?

Yes, and you should. You can swap the chat entry point by keeping the “Chat Message Trigger” but routing messages from Slack, Telegram, or a website widget into it. You can also change what gets retrieved by editing the Google Sheets retrieval step (the “Retrieve Sheet Data / Fetch Sheet Context” nodes) and the Pinecone lookup used by the “Question and Answer Chain.” Common customizations include adding a “refund policy” sheet tab, forcing citations in the AI reply, and logging extra fields like customer email or plan type.

Why is my Google Sheets connection failing in this workflow?

Usually it’s expired OAuth access or the Google account doesn’t have edit permission on the target sheet. Reconnect the Google Sheets credential in n8n, then open the sheet and confirm sharing is correct. If it fails only on busy days, you may be hitting Google API quotas and should batch writes or reduce per-message lookups.

What volume can this Sheets Drive automation workflow process?

On a typical n8n Cloud plan, it can handle hundreds to thousands of chats a month depending on your execution limits, and self-hosting removes the execution cap (your server becomes the limit). In practice, the bottlenecks are OpenAI response time and Pinecone query latency, not n8n itself.

Is this Sheets Drive automation automation better than using Zapier or Make?

For RAG-style support replies, usually yes. n8n is built for multi-step logic, branching, and “agent” style workflows that pull from multiple sources in one run, and you can self-host when volume grows. Zapier and Make can work, but you often end up stitching together lots of small automations, and costs climb once you’re logging every message and doing document ingestion. If you just want to log chats to a sheet, those tools can be quicker to start. If you want the AI to answer from Drive plus keep weekly reporting, n8n fits better. Talk to an automation expert if you’re not sure which fits.

Once this is running, your team stops reinventing answers all day. The workflow handles the repetitive lookup, drafting, and logging so you can focus on the tricky cases.

Need Help Setting This Up?

Our automation experts can build and customize this workflow for your specific needs. Free 15-minute consultation—no commitment required.

Lisa Granqvist

Workflow Automation Expert

Expert in workflow automation and no-code tools.

{
"id": null,
"meta": {
"instanceId": "0a43e272b50bc49eabb6779679f3ec844092fd1478e9fc0cd3a422b65cb86119",
"templateId": null,
"templateCredsSetupCompleted": null
},
"name": "Automated Chat Response and Logging Flow",
"tags": [],
"nodes": [
{
"id": "flowpast-topbar-10828",
"name": "Flowpast Branding",
"type": "n8n-nodes-base.stickyNote",
"position": [
0,
110
],
"parameters": {
"color": 7,
"width": 1395,
"height": 80,
"content": "## Flowpast.com | Automation Workflow Library\n**\ud83d\udcd6 Full tutorial & setup guide:** flowpast.com"
},
"typeVersion": 1
},
{
"id": "b3752497-5f33-4afb-88f9-7dcb1670f364",
"name": "Session Memory Buffer",
"type": "@n8n/n8n-nodes-langchain.memoryBufferWindow",
"position": [
560,
430
],
"parameters": {
"sessionKey": "chat-rag-session",
"sessionIdType": "customKey",
"contextWindowLength": 7
},
"typeVersion": 1.3
},
{
"id": "07e78867-bb70-43d4-9229-636f9a8d6193",
"name": "Chat Response Orchestrator",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
540,
260
],
"parameters": {
"text": "=User message: {{ $json.content }}\nIntent: {{ $json.intent }}\nTopic: {{ $json.topic || \"general\" }}\nContext (from Google Sheets or memory): {{ $json.context || $memory || \"No context retrieved\" }}\n",
"options": {
"systemMessage": "=You are \"Content Agent Chat Responder\", a professional AI assistant inside an n8n automation system.\n\nYour role:\n- Communicate naturally, professionally, and concisely.\n- Use provided context from Google Sheets or memory when available.\n- If the user greets you (like \"hi\", \"hello\", \"hey\"), respond politely and ask how you can help.\n- If the user asks about SEO, content drafts, briefs, optimization, or publishing, respond specifically based on context or ask clarifying questions if missing.\n- If the intent is \"chat\" but the content clearly mentions a topic, treat it as a mixed chat + content query.\n- If context is missing, ask the user for more information. \n\n\n### Behavior Rules:\n- Always greet politely and warmly.\n- If the user asks about our company, products, or projects, **use the context retrieved from Pinecone (knowledge base) as your source of truth**.\n- Give clear, structured, and easy-to-understand answers \u2014 avoid long walls of text.\n- When appropriate, format answers into short paragraphs, bullet points, or step-by-step explanations.\n- Sound confident, approachable, and \"sales-ready\" \u2014 as if you are introducing the company to a client.\n- If the knowledge base does not contain the answer, respond naturally but say you will get back with more details.\n- Maintain a human-like flow: use follow-up questions, confirmations, and conversational phrases (e.g., \"That's a great question!\" or \"Let me explain in simple terms\u2026\").\n\n### Example Scenarios:\n- If a client asks: \"What is this project about?\" \u2192 Provide a short overview from Pinecone, structured like a pitch.\n- If they ask: \"Can you explain in detail?\" \u2192 Give a more thorough breakdown in a clear format.\n- If they greet or ask something general \u2192 continue the chat naturally, like a real human. \n\n- Always output concise structured JSON in the format:\n{\n  \"reply\": \"Your full response text\",\n  \"context_used\": [\"list of any context used or none\"]\n}\n"
},
"promptType": "define",
"hasOutputParser": true
},
"typeVersion": 2.1
},
{
"id": "2e25a1fd-88b8-44e8-bde1-17acd49729cd",
"name": "Structured JSON Parser",
"type": "@n8n/n8n-nodes-langchain.outputParserStructured",
"position": [
750,
360
],
"parameters": {
"jsonSchemaExample": "{\n  \"reply\": \"string\",\n  \"context_used\": [\"string\"]\n}\n"
},
"typeVersion": 1.3
},
{
"id": "68c48bd8-ff0b-4c94-b9f7-2c3fb9def1d5",
"name": "OpenAI Chat Engine",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
380,
365
],
"parameters": {
"model": {
"__rl": true,
"mode": "id",
"value": "="
},
"options": []
},
"credentials": {
"openAiApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1.2
},
{
"id": "6cc6c210-636a-46ad-8e1a-cc3292639d6e",
"name": "Chat Message Trigger",
"type": "@n8n/n8n-nodes-langchain.chatTrigger",
"position": [
30,
210
],
"webhookId": "599e92f2-e435-4937-8459-d270656f689f",
"parameters": {
"options": []
},
"typeVersion": 1.3
},
{
"id": "d686aa69-9e35-43b7-9f89-64690528db5b",
"name": "Drive File Created Trigger",
"type": "n8n-nodes-base.googleDriveTrigger",
"position": [
220,
980
],
"parameters": {
"event": "fileCreated",
"options": [],
"pollTimes": {
"item": [
{
"mode": "everyMinute"
}
]
},
"triggerOn": "specificFolder",
"folderToWatch": {
"__rl": true,
"mode": "id",
"value": "="
}
},
"credentials": {
"googleDriveOAuth2Api": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1
},
{
"id": "954d129c-2d36-4ff1-8474-3b2f3384c8d5",
"name": "Retrieve Drive File",
"type": "n8n-nodes-base.googleDrive",
"position": [
470,
1010
],
"parameters": {
"fileId": {
"__rl": true,
"mode": "id",
"value": "={{ $json.id }}"
},
"options": [],
"operation": "download"
},
"credentials": {
"googleDriveOAuth2Api": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 3
},
{
"id": "c763f13b-0e7f-4abe-ae0d-f029ab45c8e2",
"name": "Standard Document Loader",
"type": "@n8n/n8n-nodes-langchain.documentDefaultDataLoader",
"position": [
760,
1205
],
"parameters": {
"options": {
"metadata": {
"metadataValues": [
{
"name": "file-name",
"value": "={{ $('Retrieve Drive File').item.json.name }}"
}
]
}
},
"dataType": "binary",
"textSplittingMode": "custom"
},
"typeVersion": 1.1
},
{
"id": "3f237c12-20f4-4e77-a991-f231875dad10",
"name": "Recursive Text Chunker",
"type": "@n8n/n8n-nodes-langchain.textSplitterRecursiveCharacterTextSplitter",
"position": [
520,
1185
],
"parameters": {
"options": [],
"chunkSize": 500,
"chunkOverlap": 200
},
"typeVersion": 1
},
{
"id": "14e92af7-eeb2-4afa-9543-62c87ba2074f",
"name": "OpenAI Embedding Builder",
"type": "@n8n/n8n-nodes-langchain.embeddingsOpenAi",
"position": [
980,
1060
],
"parameters": {
"options": {
"dimensions": 512
}
},
"credentials": {
"openAiApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1.2
},
{
"id": "fccdff58-7a8a-44a8-9a1c-a20d681b0438",
"name": "Weekly Schedule Trigger",
"type": "n8n-nodes-base.scheduleTrigger",
"position": [
0,
720
],
"parameters": {
"rule": {
"interval": [
{
"field": "weeks",
"triggerAtDay": [
1
],
"triggerAtHour": 22
}
]
}
},
"typeVersion": 1.2
},
{
"id": "b357b118-d803-4c9d-ae3d-324eee2cca08",
"name": "Combine Sheet Records",
"type": "n8n-nodes-base.aggregate",
"position": [
515,
730
],
"parameters": {
"options": [],
"aggregate": "aggregateAllItemData"
},
"typeVersion": 1
},
{
"id": "8ad58732-4dec-4649-a1c6-0b9a6b08eb6a",
"name": "Fetch Sheet Context",
"type": "n8n-nodes-base.googleSheetsTool",
"position": [
360,
530
],
"parameters": {
"sheetName": {
"__rl": true,
"mode": "list",
"value": "",
"cachedResultUrl": "",
"cachedResultName": ""
},
"documentId": {
"__rl": true,
"mode": "id",
"value": "="
}
},
"credentials": {
"googleSheetsOAuth2Api": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 4.7
},
{
"id": "8add9b4b-4086-4a8c-be4e-47c591c864ea",
"name": "Append Conversation Log",
"type": "n8n-nodes-base.googleSheets",
"position": [
820,
235
],
"parameters": {
"operation": "appendOrUpdate",
"sheetName": {
"__rl": true,
"mode": "id",
"value": ""
},
"documentId": {
"__rl": true,
"mode": "id",
"value": "="
}
},
"credentials": {
"googleSheetsOAuth2Api": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 4.7
},
{
"id": "ef7c257c-accd-47db-a799-21b9f38a933c",
"name": "Pinecone Index Upsert",
"type": "@n8n/n8n-nodes-langchain.vectorStorePinecone",
"position": [
740,
995
],
"parameters": {
"mode": "insert",
"options": [],
"pineconeIndex": {
"__rl": true,
"mode": "id",
"value": "="
},
"embeddingBatchSize": 1000
},
"credentials": {
"pineconeApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1.3
},
{
"id": "3e139499-ba6e-465d-b464-12d74fa0884f",
"name": "Pinecone Knowledge Lookup",
"type": "@n8n/n8n-nodes-langchain.vectorStorePinecone",
"position": [
945,
545
],
"parameters": {
"mode": "retrieve-as-tool",
"topK": 5,
"options": [],
"pineconeIndex": {
"__rl": true,
"mode": "list",
"value": "[YOUR_ID]",
"cachedResultName": "[YOUR_ID]"
},
"toolDescription": "=Retrieve data from the Pinecone knowledge base and use it to answer user queries about the company, products, or projects in a well-structured, human-like manner"
},
"credentials": {
"pineconeApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1.3
},
{
"id": "c3d32f3d-4804-49dd-8ee2-7da94bea33e0",
"name": "Chat Query Embeddings",
"type": "@n8n/n8n-nodes-langchain.embeddingsOpenAi",
"position": [
710,
560
],
"parameters": {
"options": {
"dimensions": 512
}
},
"credentials": {
"openAiApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1.2
},
{
"id": "83b6174f-12c9-440f-a021-7703570a3b48",
"name": "Retrieve Sheet Data",
"type": "n8n-nodes-base.googleSheets",
"position": [
245,
690
],
"parameters": {
"sheetName": {
"__rl": true,
"mode": "list",
"value": "",
"cachedResultUrl": "",
"cachedResultName": ""
},
"documentId": {
"__rl": true,
"mode": "id",
"value": "="
}
},
"credentials": {
"googleSheetsOAuth2Api": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 4.7
},
{
"id": "561655e9-9810-4cd4-a770-794d7a60f6cb",
"name": "Convert Records to File",
"type": "n8n-nodes-base.convertToFile",
"position": [
785,
700
],
"parameters": {
"options": [],
"operation": "toText",
"sourceProperty": "data"
},
"typeVersion": 1.1
},
{
"id": "5f10e45a-6910-4cc3-897b-ba198d43943d",
"name": "Validate File Type",
"type": "n8n-nodes-base.if",
"position": [
1045,
735
],
"parameters": {
"options": [],
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "a10217a0-9aa9-47eb-9179-b1760ae4fbd4",
"operator": {
"name": "filter.operator.equals",
"type": "string",
"operation": "equals"
},
"leftValue": "={{ $binary.data.mimeType }}",
"rightValue": "application/json"
}
]
}
},
"typeVersion": 2.2
},
{
"id": "e1faaf7b-1b72-460d-94fa-dbdb719e9256",
"name": "Email History Attachment",
"type": "n8n-nodes-base.gmail",
"position": [
1295,
705
],
"webhookId": "9f4a4d22-8f33-4814-af35-7c5c916e9b2c",
"parameters": {
"message": "=Hi [Name], \n\nPlease find attached your chat history from our conversation on chat_history_{{ $now.format('DD') }}.   Feel free to refer back to this anytime, and let me know if you have any questions! \n\nBest regards, [Your Name]",
"options": {
"attachmentsUi": {
"attachmentsBinary": [
[]
]
}
},
"subject": "=chat_history_{{ $now.format('DD') }}",
"emailType": "text"
},
"credentials": {
"gmailOAuth2": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 2.1
},
{
"id": "8a43b13d-92ea-4847-8080-fa1a8b373253",
"name": "Prepare Agent Inputs",
"type": "n8n-nodes-base.set",
"position": [
255,
230
],
"parameters": {
"options": [],
"assignments": {
"assignments": [
{
"id": "d917c6d3-ae35-4140-99c8-ae390866e735",
"name": "intent",
"type": "string",
"value": "=Chat"
},
{
"id": "e697ec89-26ed-4fb3-95a9-4a0cc6eed36d",
"name": "topic",
"type": "string",
"value": "=AI Seo Basics"
},
{
"id": "f5a88ad2-7f57-4971-88ea-bad9e1c10992",
"name": "content_id",
"type": "string",
"value": "=C001"
},
{
"id": "06fb1d17-f233-4b22-af27-cd3351c971a0",
"name": "parameter",
"type": "object",
"value": "={{ $json.parameter }}"
}
]
}
},
"typeVersion": 3.4
}
],
"active": false,
"pinData": [],
"settings": {
"executionOrder": "v1"
},
"versionId": null,
"connections": {
"Combine Sheet Records": {
"main": [
[
{
"node": "Convert Records to File",
"type": "main",
"index": 0
}
]
]
},
"Retrieve Drive File": {
"main": [
[
{
"node": "Pinecone Index Upsert",
"type": "main",
"index": 0
}
]
]
},
"Validate File Type": {
"main": [
[
{
"node": "Email History Attachment",
"type": "main",
"index": 0
}
]
]
},
"Weekly Schedule Trigger": {
"main": [
[
{
"node": "Retrieve Sheet Data",
"type": "main",
"index": 0
}
]
]
},
"OpenAI Embedding Builder": {
"ai_embedding": [
[
{
"node": "Pinecone Index Upsert",
"type": "ai_embedding",
"index": 0
}
]
]
},
"OpenAI Chat Engine": {
"ai_languageModel": [
[
{
"node": "Chat Response Orchestrator",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Session Memory Buffer": {
"ai_memory": [
[
{
"node": "Chat Response Orchestrator",
"type": "ai_memory",
"index": 0
}
]
]
},
"Standard Document Loader": {
"ai_document": [
[
{
"node": "Pinecone Index Upsert",
"type": "ai_document",
"index": 0
}
]
]
},
"Retrieve Sheet Data": {
"main": [
[
{
"node": "Combine Sheet Records",
"type": "main",
"index": 0
}
]
]
},
"Drive File Created Trigger": {
"main": [
[
{
"node": "Retrieve Drive File",
"type": "main",
"index": 0
}
]
]
},
"Convert Records to File": {
"main": [
[
{
"node": "Validate File Type",
"type": "main",
"index": 0
}
]
]
},
"Chat Response Orchestrator": {
"main": [
[
{
"node": "Append Conversation Log",
"type": "main",
"index": 0
}
]
]
},
"Prepare Agent Inputs": {
"main": [
[
{
"node": "Chat Response Orchestrator",
"type": "main",
"index": 0
}
]
]
},
"Chat Message Trigger": {
"main": [
[
{
"node": "Prepare Agent Inputs",
"type": "main",
"index": 0
}
]
]
},
"Fetch Sheet Context": {
"ai_tool": [
[
{
"node": "Chat Response Orchestrator",
"type": "ai_tool",
"index": 0
}
]
]
},
"Structured JSON Parser": {
"ai_outputParser": [
[
{
"node": "Chat Response Orchestrator",
"type": "ai_outputParser",
"index": 0
}
]
]
},
"Recursive Text Chunker": {
"ai_textSplitter": [
[
{
"node": "Standard Document Loader",
"type": "ai_textSplitter",
"index": 0
}
]
]
},
"Chat Query Embeddings": {
"ai_embedding": [
[
{
"node": "Pinecone Knowledge Lookup",
"type": "ai_embedding",
"index": 0
}
]
]
},
"Pinecone Knowledge Lookup": {
"ai_tool": [
[
{
"node": "Chat Response Orchestrator",
"type": "ai_tool",
"index": 0
}
]
]
}
}
}