Automate AI-Powered Email Classification and Smart Labeling

This n8n workflow automates the process of classifying and labeling emails using AI. It's designed to help you organize your Gmail inbox efficiently by understanding email content and applying appropriate labels.

What it does

This workflow performs the following key steps:

1. Manual Trigger: The workflow is initiated manually, allowing you to control when the email classification process begins.
2. HTTP Request (Fetch Emails): It makes an HTTP request to an external API (likely a custom endpoint or a service that retrieves emails, given the directory name) to fetch email data.
3. Loop Over Items (Process Emails Individually): Each fetched email item is processed individually in a loop.
4. Edit Fields (Prepare Email Content): Within the loop, the workflow prepares the email content for AI processing by setting up the necessary fields.
5. Basic LLM Chain (AI Classification): The prepared email content is sent to a Language Model (LLM) chain for classification.
6. OpenAI Chat Model (AI Engine): The LLM chain leverages an OpenAI Chat Model to analyze the email content and determine its category or intent.
7. Structured Output Parser (Extract Classification): The AI's response is then parsed to extract the structured classification (e.g., category, priority, sentiment).
8. Switch (Route based on Classification): Based on the AI's classification, the workflow uses a Switch node to route the email to different paths for further action (e.g., applying specific labels, sending notifications).
9. Wordpress (Example Action): One of the potential branches from the Switch node leads to a Wordpress node, suggesting that classified emails could trigger actions like creating a post or updating content on a Wordpress site. This is an example of a possible action based on classification.
10. Aggregate (Combine Results): After processing individual emails, the results are aggregated.
11. Split Out (Further Processing): The aggregated results are then split out, likely for subsequent actions or reporting.

Prerequisites/Requirements

To use this workflow, you will need:

• n8n Instance: A running instance of n8n.
• OpenAI API Key: For the "OpenAI Chat Model" node to function, you'll need an OpenAI API key configured as a credential in n8n.
• Gmail Account (Implicit): While not directly configured in the provided JSON, the workflow's purpose (email classification and labeling) strongly implies interaction with a Gmail account or a similar email service, likely through the initial HTTP Request.
• External Email Fetching API/Service: The "HTTP Request" node suggests an external API or service is used to retrieve emails. You'll need access to and configuration details for this service.
• Wordpress Credentials (Optional): If you intend to use the Wordpress branch of the workflow, you will need a Wordpress account and corresponding n8n credentials.

Setup/Usage

1. Import the Workflow: Download the provided JSON and import it into your n8n instance.
2. Configure Credentials:
   • Set up your OpenAI API Key credential in n8n.
   • Configure any necessary credentials for the HTTP Request node if it requires authentication to fetch emails.
   • If using the Wordpress node, configure your Wordpress credentials.
3. Customize HTTP Request: Adjust the "HTTP Request" node to point to your email fetching API/service and include any required parameters for retrieving emails.
4. Refine AI Classification:
   • Review the "Basic LLM Chain" and "OpenAI Chat Model" configurations. You may need to customize the prompt or model settings to achieve the desired classification accuracy for your specific email types.
   • Adjust the "Structured Output Parser" to correctly extract the classification data based on the AI's output format.
5. Configure Switch Node: Customize the "Switch" node's conditions to define how emails should be routed based on the AI's classification (e.g., if category is "Marketing", then...).
6. Define Actions:
   • Modify or add nodes after the "Switch" node to perform specific actions based on the classification (e.g., add a Gmail label, send a Slack notification, update a CRM). The "Wordpress" node is an example; you can replace or augment it with other actions.
7. Activate and Execute: Once configured, activate the workflow. You can then execute it manually using the "When clicking ‘Execute workflow’" trigger.