Contextual n8n Splitting

This workflow is for anyone who needs to process and store text data from n8n workflows into a Pinecone vector database. It is especially useful for developers and data engineers working with RAG (Retrieval Augmented Generation) systems. This workflow extracts workflow data, splits it into manageable chunks, and uses AI to add context to each chunk. It then stores these contextualized chunks in a Pinecone vector database for efficient retrieval. You must connect your n8n instance to Airtable, Pinecone, and OpenAI. You will also need to define the Pinecone index and ensure your Airtable base contains the necessary workflow information. You need an n8n instance, accounts for Airtable, Pinecone, and OpenAI, and a basic understanding of n8n workflows and API keys. You can customize the text splitting parameters, the AI prompt for contextualization, and the Airtable base/table used for storing and tracking workflows. Key Technologies: n8n,Airtable,Pinecone,OpenAI (GPT-4.1-mini, text-embedding-3-large) Automation Value: - Automatically prepares n8n workflow data for RAG systems. - Ensures data is properly chunked and contextualized, improving searchability. - Reduces manual effort in preparing data for vector databases. - Integrates with Airtable to track which workflows have been processed. Best Practices: Ensure consistent naming conventions for nodes and variables to improve workflow readability and maintenance. Implement reliable error handling for API calls (e.g., Pinecone, OpenAI) to manage rate limits or service outages gracefully. Regularly monitor the performance and cost of AI models to optimize usage and budget. Document the AI model's prompt and expected output format thoroughly to ensure consistent contextualization. Consider using environment variables for sensitive credentials and frequently used parameters to enhance security and reusability.