This n8n workflow automates the process of backing up transaction logs, analyzing their content, and storing key information for future reference. It is triggered by an HTTP POST request sent to a specific webhook, making it suitable for real-time data processing from transactional systems.
The workflow begins with a webhook trigger that starts the process when transaction data is received. The data is then split into manageable chunks by the Text Splitter node, which uses character-based splitting. Each chunk is transformed into embeddings using the Cohere API, enabling semantic understanding of the logs.
These embeddings are stored in a Supabase vector database, which facilitates efficient similarity searches and analysis later in the process. The workflow then queries this database to retrieve relevant log data and analyzes it further using a language model (OpenAI) within a RAG (Retrieval-Augmented Generation) agent. The RAG agent is configured to handle the transaction log data as per the defined prompt.
Based on the analysis, the workflow appends results or status updates into a Google Sheet for record-keeping. If any errors occur during processing, a Slack alert is triggered to notify the administrator, ensuring prompt attention.
This workflow can be used by financial or transactional systems to automate log backups, perform semantic analysis of logs for security or troubleshooting, and maintain an organized record of transaction processing events.
Reviews
There are no reviews yet.