Leon van Zyl writes:
Learn how to implement long-term memory in your AI agents using n8n! This tutorial shows you how to create AI assistants that remember user preferences across multiple conversations - similar to ChatGPT’s memory feature.
2 Likes
Hi there!
Im new to n8n. I tryied to build an Airtable tool for my ai agent using your video. But unfortunatelly the AI agent dont use it at all during execution. What can be the problem? I could not find the download link of your workflow. Cheers!
I don’t know if you ever got to work around this
This is a common agent setup issue rather than an Airtable problem.
In n8n, the AI Agent will only use tools that are clearly exposed and scoped. Double-check that your Airtable node is connected as a tool with a very explicit name and description. If the description is vague, the agent usually ignores it.
Also look at the agent prompt. If the task can be completed without querying Airtable, the agent won’t call the tool at all. The prompt needs to force a dependency on Airtable data.
If you’re using a sub-workflow, make sure it’s marked as “Expose as tool.” That’s an easy one to miss.
As for the workflow file, a lot of tutorial videos don’t include the JSON unless it’s linked in the description or comments. If there’s no link, you’ll probably need to rebuild it or ask the creator to share it.