Extract Email Tasks with Gmail, ChatGPT-4o, and Supabase

This n8n workflow automates the process of identifying and extracting actionable tasks from your Gmail inbox, processing them with OpenAI's ChatGPT-4o, and storing them in a Supabase database. It helps you keep track of tasks embedded in your emails without manual sifting.

What it does

1. Triggers on a Schedule: The workflow runs at a predefined interval (e.g., daily, hourly) to check for new emails.
2. Fetches Unread Gmail Messages: It connects to your Gmail account and retrieves unread emails.
3. Filters for Relevant Emails: An If node checks if the email subject or body contains keywords like "task", "todo", or "action item" to narrow down the relevant messages.
4. Extracts Tasks with OpenAI (ChatGPT-4o): For each filtered email, it sends the email content to OpenAI's ChatGPT-4o model, prompting it to identify and extract any explicit or implicit tasks.
5. Processes OpenAI Response: A Code node then parses the response from OpenAI, extracting the structured task information.
6. Stores Tasks in Supabase: The extracted tasks are then inserted into a Supabase database, allowing for centralized task management.
7. Marks Emails as Read (Optional): After processing, the workflow can mark the emails as read in Gmail to avoid reprocessing.

Prerequisites/Requirements

To use this workflow, you will need:

• n8n Account: A running instance of n8n (cloud or self-hosted).
• Gmail Account: A Google account with Gmail access and credentials configured in n8n.
• OpenAI API Key: An OpenAI API key with access to the gpt-4o model, configured as a credential in n8n.
• Supabase Account: A Supabase project with a database table set up to store your tasks, and credentials configured in n8n.
  • The table should ideally have columns for task_description, email_subject, email_id, due_date (if extractable), priority, etc.

Setup/Usage

1. Import the Workflow:
   • Download the provided JSON file.
   • In your n8n instance, go to "Workflows" and click "New".
   • Click the "Import from JSON" button and paste the workflow JSON or upload the file.
2. Configure Credentials:
   • Locate the Gmail node and configure your Google OAuth2 credentials.
   • Locate the OpenAI node and configure your OpenAI API Key credentials.
   • Locate the Supabase node and configure your Supabase API credentials (Project URL and API Key).
3. Customize Nodes:
   • Schedule Trigger: Adjust the schedule to your preferred frequency (e.g., every 30 minutes, once a day).
   • Gmail: You might want to adjust the "Label" or "Query" parameters to refine which emails are fetched (e.g., is:unread in:inbox category:primary).
   • If: Refine the conditions in the If node to better filter emails relevant to tasks based on your specific needs.
   • OpenAI: Review the prompt used to extract tasks. You might want to modify it to include specific instructions for task format or additional details you want to extract (e.g., "Extract tasks, due dates, and assignees in JSON format.").
   • Code: If you modify the OpenAI prompt to return a different JSON structure, you will need to update the Code node to correctly parse the new format.
   • Supabase: Ensure the table name and column mappings in the Supabase node match your Supabase database schema.
4. Activate the Workflow: Once configured, activate the workflow by toggling the "Active" switch in the top right corner of the workflow editor.

The workflow will now automatically run on its schedule, process your emails, and populate your Supabase database with extracted tasks.