Automated customer support with GPT-4 knowledge base agent for Gmail
Ready for 3.0What this workflow does
This AI workflow has 16 nodes and runs on Google Drive events (file created) (plus 1 other trigger). Along the way it connects OpenAI Chat Model, Embeddings OpenAI, Pinecone Vector Store and Gmail and 1 more.
How it runs
Trigger: On Google Drive events (file created) (plus 1 other trigger).
- Download file (Google Drive): download
- Text Classifier: classifies text into categories with AI, with model: OpenAI Chat Model
- Pinecone Vector Store: stores documents in the Pinecone vector store, with embeddings: Embeddings OpenAI; document loader: Default Data Loader
- AI Agent: runs an AI agent, with model: OpenAI Chat Model; 1 tool (knowledgebase)
- Add label to message (Gmail): add labels
- Reply to a message (Gmail): reply
Services it uses
Before you upgrade to n8n 3.0
Nothing in this workflow is removed in n8n 3.0, so it should import and run after the upgrade.
- Still check Gmail Trigger below v1.4 runs as v1.4 (drafts skipped, max emails per poll) (Gmail Trigger).
Nodes used (12)
- Text Classifier
@n8n/n8n-nodes-langchain.textClassifier - OpenAI Chat Model
@n8n/n8n-nodes-langchain.lmChatOpenAi - AI Agent
@n8n/n8n-nodes-langchain.agent - Embeddings OpenAI
@n8n/n8n-nodes-langchain.embeddingsOpenAi - Pinecone Vector Store
@n8n/n8n-nodes-langchain.vectorStorePinecone - Gmail
n8n-nodes-base.gmail - No Operation, do nothing
n8n-nodes-base.noOp - Google Drive Trigger
n8n-nodes-base.googleDriveTrigger - Google Drive
n8n-nodes-base.googleDrive - Default Data Loader
@n8n/n8n-nodes-langchain.documentDefaultDataLoader - Recursive Character Text Splitter
@n8n/n8n-nodes-langchain.textSplitterRecursiveCharacterTextSplitter - Gmail Trigger
n8n-nodes-base.gmailTrigger
From the author
Customer Support AI Agent for Gmail This n8n template demonstrates how to build an AI-powered customer support workflow that automatically handles incoming Gmail messages, classifies them, finds answers from your knowledge base, and sends a personalized…