AI Agent for Q&A on n8n Workflow Data using LangChain & OpenAI

Unlock Dynamic Insights from Your Existing Automations with this AI Agent

This n8n workflow acts as an intelligent AI Agent capable of querying the data generated by any other n8n workflow. You specify a target workflow ID, and this agent retrieves its output. Then, using LangChain's RetrievalQAChain powered by an OpenAI chat model, it answers your natural language questions based on that retrieved data. Imagine having a direct line to ask, 'What were the key outcomes of my lead generation workflow last week?' or 'Find all customer support tickets related to billing from yesterday's run.' This agent makes that possible, transforming static workflow outputs into interactive knowledge bases.

Key Features & Benefits

• Cross-Workflow Intelligence: Query data from any of your n8n workflows by simply providing its ID.
• Natural Language Q&A: Ask questions in plain English and get answers directly from your workflow's data.
• LangChain & OpenAI Powered: Leverages sophisticated LangChain RetrievalQAChain and OpenAI's advanced chat models for accurate understanding and response generation.
• Dynamic Data Exploration: No more manual sifting through JSON outputs or CSVs. Get targeted information quickly.
• Boosts Automation ROI: Extract more value and actionable insights from the data your automated processes already generate.
• Easy to Configure: Set your target workflow ID, OpenAI credentials, and your query to get started.

• An n8n instance (Cloud or self-hosted, with LangChain nodes available/installed).
• OpenAI API Key with access to a suitable chat model (e.g., gpt-3.5-turbo or gpt-4).
• The ID of an existing n8n workflow whose output data you want to query.

1. Download the n8n workflow JSON file.
2. Import the workflow into your n8n instance.
3. Configure the 'OpenAI Chat Model' node: select your OpenAI credential and choose your desired model.
4. In the 'Workflow Retriever' node, enter the Workflow ID for the target n8n workflow whose data you wish to query. (The workflow's internal sticky note refers to this as a 'Subworkflow ID').
5. Customize the 'Example Prompt' node's input value with the question you want to ask about the target workflow's data.
6. Ensure all necessary nodes are connected as per the workflow structure.
7. Activate the workflow and run it manually using the 'When clicking "Execute Workflow"' trigger to test.