Extract Tasks from Gmail with GPT-4o & Supabase

This n8n workflow triggers on a schedule to fetch unread emails from Gmail, processes each one individually via SplitInBatches to avoid overload, and checks a Supabase 'emails' table to prevent re-processing using a unique email_id constraint.

For new emails, it formats the content into a structured prompt and sends it to OpenAI's GPT-4o model to extract task metadata like summaries, priorities, and actions. The AI-generated structured data is then inserted into Supabase via HTTP REST API, enabling seamless integration with task management systems.

Benefits include time-saving automation for busy professionals, reducing manual email triage, and centralizing task data in a database for querying or further automation. Use cases: sales teams extracting leads from inquiries, project managers capturing action items from vendor emails, or personal productivity setups turning inbox chaos into organized tasks.