This n8n workflow automates the process of extracting personal data from user messages using a self-hosted language model (Mistral NeMo) integrated via Ollama. The workflow is triggered when a chat message is received, and it leverages multiple nodes for language processing, data extraction, auto-correction, and structured output formatting.

First, a webhook listens for incoming chat messages. Upon receiving a message, the data is processed by a language model node configured with the Mistral NeMo model, which generates a response based on the user input with a low temperature setting for more deterministic output. The response goes through an auto-fixing parser if it doesn’t meet the predefined schema, prompting the model to correct its output.

A structured output parser then verifies if the response aligns with the expected JSON schema—containing fields like name, surname, contact details, timestamp, and subject. If the output passes validation, it is extracted and stored. In case of errors, a no-op node handles the failure gracefully.

This workflow is useful for automating the extraction of structured personal data from chat interactions, suitable for customer service, form automation, or onboarding processes where data consistency and accuracy are critical.

Node Count	11 – 20 Nodes
Nodes Used	@n8n/n8n-nodes-langchain.chainLlm, @n8n/n8n-nodes-langchain.chatTrigger, @n8n/n8n-nodes-langchain.lmChatOllama, @n8n/n8n-nodes-langchain.outputParserAutofixing, @n8n/n8n-nodes-langchain.outputParserStructured, noOp, set, stickyNote