Parse Invoice Emails Using RAG Agent and Weaviate

Overview

This n8n workflow automates the process of parsing invoice emails. It utilizes a Retrieval Augmented Generation (RAG) agent to intelligently extract information from the email content. The workflow begins with a Webhook Trigger to receive incoming emails. The email content is then split into manageable chunks using the Text Splitter node. These chunks are converted into embeddings using the Embeddings node and stored in Weaviate, a vector database, for efficient retrieval. When a new email arrives, the Weaviate Query node retrieves relevant context, which is then fed to the RAG Agent along with the email content and chat history (managed by Window Memory) to generate a structured output. The results are logged to a Google Sheet via the Append Sheet node, and any errors are reported via Slack Alert.