Document-based chatbot with memory using OpenAI, Pinecone and Google Drive

Open on n8n.io

$20/month : Unlimited workflows

2500 executions/month

Try free

THE #1 IN WEB SCRAPING

Scrape any website without limits

Try free

HOSTINGER

Early Deal
DISCOUNT 20%

Self-hosted n8n

Unlimited workflows - from $4.99/mo

Try free

#1 hub for scraping, AI & automation

6000+ actors - $5 credits/mo

Try free

Important notice

This workflow is provided as-is. Please review and test before using in production.

1. Workflow Overview

Who is This For? This is for normal people or people just starting off and wanting to have a AI chatbot that can process data to use when talking to the user. How to Use You will need to have your ...

Best for

Personal Productivity automation workflows
AI RAG automation workflows
advanced n8n builders looking for reusable templates

Tools used

@n8n/n8n-nodes-langchain.chattrigger, @n8n/n8n-nodes-langchain.agent, n8n-nodes-base.airtabletool, n8n-nodes-base.airtable, n8n-nodes-base.aggregate, n8n-nodes-base.merge, @n8n/n8n-nodes-langchain.memorybufferwindow, n8n-nodes-base.manualtrigger

Source and attribution

This workflow is cataloged by N8N Workflows and links back to its original n8n.io source page by Sally.

Original n8n.io source

1.1 Workflow description

Title: Document-based chatbot with memory using OpenAI, Pinecone and Google Drive
Workflow name: Document-based chatbot with memory using OpenAI, Pinecone and Google Drive

Who is This For? This is for normal people or people just starting off and wanting to have a AI chatbot that can process data to use when talking to the user.

How to Use You will need to have your own OpenRouter (Free) and OpenAI APIs as well as Google Drive, Pinecone, and Airtable.

What Do You Want? If you want to have your AI Agent remember the user's preferences even after the session is over then you can keep the Airtable node in, if not you can delete it.

1.2 Logical Blocks

This catalog entry is organized from the workflow JSON. The node-level section below shows the executable blocks available for review before importing the template.

2. Block-by-Block Analysis

Block 1 - When chat message received

Type / Role: @n8n/n8n-nodes-langchain.chatTrigger - chatTrigger
Config choices: Version 1.1

Block 2 - AI Agent

Type / Role: @n8n/n8n-nodes-langchain.agent - agent
Config choices: Version 2

Block 3 - Save Memory

Type / Role: n8n-nodes-base.airtableTool - airtableTool
Config choices: Version 2.1

Block 4 - Get Memories

Type / Role: n8n-nodes-base.airtable - airtable
Config choices: Version 2.1

Block 5 - Aggregate

Type / Role: n8n-nodes-base.aggregate - aggregate
Config choices: Version 1

Block 6 - Merge

Type / Role: n8n-nodes-base.merge - merge
Config choices: Version 3.2

Block 7 - Simple Memory

Type / Role: @n8n/n8n-nodes-langchain.memoryBufferWindow - memoryBufferWindow
Config choices: Version 1.3

Block 8 - When clicking 'Test Workflow' button

Type / Role: n8n-nodes-base.manualTrigger - manualTrigger
Config choices: Version 1

Block 9 - Google Drive

Type / Role: n8n-nodes-base.googleDrive - googleDrive
Config choices: Version 3

Block 10 - Get Content

Type / Role: n8n-nodes-base.googleDrive - googleDrive
Config choices: Version 3

Block 11 - Loop Over Items

Type / Role: n8n-nodes-base.splitInBatches - splitInBatches
Config choices: Version 3

Block 12 - Pinecone Vector Store

Type / Role: @n8n/n8n-nodes-langchain.vectorStorePinecone - vectorStorePinecone
Config choices: Version 1.2

Block 13 - Embeddings OpenAI

Type / Role: @n8n/n8n-nodes-langchain.embeddingsOpenAi - embeddingsOpenAi
Config choices: Version 1.2

Block 14 - Default Data Loader

Type / Role: @n8n/n8n-nodes-langchain.documentDefaultDataLoader - documentDefaultDataLoader
Config choices: Version 1

Block 15 - Recursive Character Text Splitter

Type / Role: @n8n/n8n-nodes-langchain.textSplitterRecursiveCharacterTextSplitter - textSplitterRecursiveCharacterTextSplitter
Config choices: Version 1

Block 16 - Answer questions with a vector store

Type / Role: @n8n/n8n-nodes-langchain.toolVectorStore - toolVectorStore
Config choices: Version 1.1

Block 17 - OpenAI Chat Model1

Type / Role: @n8n/n8n-nodes-langchain.lmChatOpenAi - lmChatOpenAi
Config choices: Version 1.2

Block 18 - Pinecone Vector Store1

Type / Role: @n8n/n8n-nodes-langchain.vectorStorePinecone - vectorStorePinecone
Config choices: Version 1.2

Block 19 - Embeddings OpenAI1

Type / Role: @n8n/n8n-nodes-langchain.embeddingsOpenAi - embeddingsOpenAi
Config choices: Version 1.2

Block 20 - Sticky Note

Type / Role: n8n-nodes-base.stickyNote - stickyNote
Config choices: Version 1

Block 21 - Sticky Note1

Type / Role: n8n-nodes-base.stickyNote - stickyNote
Config choices: Version 1

Block 22 - OpenRouter Chat Model

Type / Role: @n8n/n8n-nodes-langchain.lmChatOpenRouter - lmChatOpenRouter
Config choices: Version 1

3. Summary Table

Workflow	Document-based chatbot with memory using OpenAI, Pinecone and Google Drive
Complexity	advanced
Nodes	22
Categories	Personal Productivity, AI RAG
Author	Sally
Published	14 Jun 2025

4. Reproducing the Workflow from Scratch

1. Download the workflow JSON

Use the JSON export at /data/workflows/4930/4930.json as the source template for this automation.
2. Import the template into n8n

Open n8n, import the downloaded JSON, and review each node before activating the workflow.
3. Configure credentials and variables

Replace placeholder credentials, API keys, webhook URLs, account IDs, and environment-specific values with your own settings.
4. Test with sample data

Run the workflow manually or in a staging workspace, inspect node output, and confirm downstream systems receive the expected data.
5. Activate and monitor

Enable the workflow only after testing, then monitor executions, errors, and rate limits during the first production runs.

5. General Notes & Resources

Review imported nodes carefully before activation. This catalog entry is intended to help you inspect the workflow structure, understand required services, and find related templates faster.

Node names, credentials, schedules, webhook paths, and external service limits may need adjustment for your workspace.

Download workflow JSON Original n8n.io source Personal Productivity workflows AI RAG workflows

Frequently asked questions

What does Document-based chatbot with memory using OpenAI, Pinecone and Google Drive do?

What do I need before importing this workflow?

Review the workflow JSON, configure any required credentials in n8n, and test the automation in a safe workspace before using it in production.

Can I customize this workflow?

Yes. Use the block-by-block analysis and the downloadable JSON to inspect each node, then adjust credentials, prompts, schedules, filters, or destinations for your Personal Productivity, AI RAG use case.

Sally

1 workflow

Nodes

@n8n/n8n-nodes-langchain.chattrigger @n8n/n8n-nodes-langchain.agent n8n-nodes-base.airtabletool n8n-nodes-base.airtable n8n-nodes-base.aggregate n8n-nodes-base.merge @n8n/n8n-nodes-langchain.memorybufferwindow n8n-nodes-base.manualtrigger

Complexity

advanced

Published 14 Jun 2025

Likes 0

View on n8n.io Download Workflow

Install path: /data/workflows/4930/4930.json

Share Your Workflow

Have a useful automation to share? Publish it and help the community.

Submit Your Template How to Submit

Related Workflows

Manage Google Calendar events with Zalo Bot, OpenAI and Gemini

## Quick Overview This workflow lets users manage Google Calendar by sending text or voice messages to a Zalo Official Account bot, transcribing voice notes with Google Gemini and using an OpenAI-powered agent to create, update, list, or delete events, then replying in Zalo with the result. ## How it works 1. Receives an incoming text or voice message through a Zalo Bot trigger. 2. Sends a typing indicator back to the Zalo chat and normalizes key message fields (chat ID, type, text, and voice URL). 3. If the message is a voice note, sends the audio URL to Google Gemini to transcribe it into text. 4. Sends the combined text/transcription to an OpenAI chat model agent that interprets the request in Vietnamese and decides whether to list, create, update, or delete a Google Calendar event. 5. Uses Google Calendar tools to get events and then create, update, or delete the targeted event based on the agent’s extracted parameters. 6. Sends the agent’s Markdown-formatted response back to the user in Zalo, and posts an error message to Zalo if the agent run fails. ## Setup 1. Install the community node package `n8n-nodes-zalo-bot-official` and restart n8n if required. 2. Connect your Zalo Bot API credentials and configure the Zalo webhook URL in your Zalo Official Account so incoming messages reach this workflow. 3. Add a Google Calendar OAuth2 credential and set the target calendar account (currently configured as `[email protected]`) for all calendar operations. 4. Add a Google Gemini (PaLM) API credential for audio transcription and ensure voice messages include a reachable `voice_url`. 5. Add an OpenAI API credential, select the chat model you want to use (currently `gpt-5-mini`), and adjust the system prompt/tone if needed.

View

Screen LinkedIn jobs and generate tailored resume and cover PDFs with Anthropic and DocRaptor

## Quick Overview This workflow collects your job preferences and resume via an n8n form, scrapes fresh LinkedIn job listings with Apify, screens and ranks matches using Anthropic Claude, generates tailored resume and cover letter PDFs via DocRaptor, then emails them to you and logs each match in Google Sheets. ## How it works 1. Receives a form submission with your email, target role, seniority, locations, skills, deal-breakers, and resume text. 2. Calls the Apify LinkedIn Job Scraper API to fetch up to 150 jobs posted in the last 7 days for the provided keywords and location. 3. Removes duplicates and filters out obvious mismatches based on seniority, required years of experience patterns, and location rules. 4. Uses Anthropic Claude (Haiku) to score each remaining job for relevance and eligibility and returns a JSON verdict and fit score. 5. Ranks the results, drops rejected roles, and keeps the top five best-scoring matches. 6. Uses Anthropic Claude (Sonnet) to generate a structured resume JSON and cover letter JSON tailored to each shortlisted job using only facts from your provided profile. 7. Renders the resume and cover letter to HTML, converts them to PDF files with DocRaptor, emails the PDFs to you via Gmail, and appends a tracker row to Google Sheets. ## Setup 1. Add credentials for Anthropic (Claude), Gmail, Google Sheets, and DocRaptor (HTTP Basic Auth with your API key as the username and a blank password). 2. Replace YOUR_APIFY_TOKEN in the Apify HTTP request URL and ensure you have access to the Apify LinkedIn Jobs Scraper actor endpoint used in the workflow. 3. Create a Google Sheet with a “Tracker” tab and set the document ID (YOUR_TRACKER_SHEET_ID) to match your sheet. 4. If you want non-watermarked PDFs, set DocRaptor "test" to false in both PDF generation requests.

View

Triage and route tasks with Google Sheets, GPT-4o-mini, Gmail, Slack, and Google Tasks

## Quick Overview This workflow collects tasks via a webhook, stores validated tasks in Google Sheets, and runs a daily prioritization routine that creates urgent items in Google Tasks with Slack alerts while summarizing medium/low tasks with OpenAI and emailing the digest via Gmail. ## How it works 1. Receives a POST webhook request containing task details and normalizes fields like priority and due date. 2. Validates required fields (including allowed priority values) and rejects invalid submissions with an error. 3. Appends valid tasks to Google Sheets with a pending status and returns a success response to the webhook caller. 4. Runs daily on a schedule, reads tasks from Google Sheets, keeps only pending items, removes duplicates, and sorts them by priority. 5. For high-priority tasks, creates a Google Tasks item and sends a Slack alert. 6. For medium and low-priority tasks, compiles a digest, generates an email-ready summary with OpenAI, and sends it via Gmail. 7. Updates each handled task in Google Sheets with a Processed status and a processed timestamp. ## Setup 1. Create a Google Sheet with columns for taskName, priority, category, dueDate, status, createdAt, and processedAt, and update the spreadsheet and sheet IDs in the Google Sheets steps. 2. Add Google Sheets OAuth2 credentials for both reading and updating/appending rows. 3. Add credentials for Google Tasks, Slack (OAuth2), Gmail, and OpenAI, and set the target Google Task list, Slack destination, and email recipients/subject as needed. 4. Copy the webhook URL from the webhook trigger and configure your task intake source to send POST requests with fields like "Task Name", "Priority", "Category", and "Due Date". 5. Adjust the schedule time and any validation rules (required fields and accepted priority values) to match your process.

View

Need Custom Automation?

Get help designing a custom n8n workflow that connects your stack and fits your process.

Document-based chatbot with memory using OpenAI, Pinecone and Google Drive

Workflow preview

Important notice

1. Workflow Overview

Best for

Tools used

Source and attribution

1.1 Workflow description

1.2 Logical Blocks

2. Block-by-Block Analysis

Block 1 - When chat message received

Block 2 - AI Agent

Block 3 - Save Memory

Block 4 - Get Memories

Block 5 - Aggregate

Block 6 - Merge

Block 7 - Simple Memory

Block 8 - When clicking 'Test Workflow' button

Block 9 - Google Drive

Block 10 - Get Content

Block 11 - Loop Over Items

Block 12 - Pinecone Vector Store

Block 13 - Embeddings OpenAI

Block 14 - Default Data Loader

Block 15 - Recursive Character Text Splitter

Block 16 - Answer questions with a vector store

Block 17 - OpenAI Chat Model1

Block 18 - Pinecone Vector Store1

Block 19 - Embeddings OpenAI1

Block 20 - Sticky Note

Block 21 - Sticky Note1

Block 22 - OpenRouter Chat Model

3. Summary Table

4. Reproducing the Workflow from Scratch

1. Download the workflow JSON

2. Import the template into n8n

3. Configure credentials and variables

4. Test with sample data

5. Activate and monitor

5. General Notes & Resources

Frequently asked questions