Chat with your data and keep Airtable up to date. A simple chat message can find, add, edit, or remove records in your base. Great for marketing teams that manage social posts, campaigns, or content ideas and want quick updates without opening Airtable.

A chat event starts an AI agent that uses OpenAI and a short memory to keep context across messages. The agent connects to a Model Context Protocol server in n8n, which exposes Airtable actions as tools. These tools handle get, search with formula filters, create, update, and delete on a set base and table. This design lets non technical users ask plain questions while the agent runs the exact Airtable steps in the background. Results return to the chat, so teams can make decisions fast.

You will need an Airtable Personal Access Token with access to your base and an OpenAI API key. Set the MCP Server Trigger path, activate it, and paste its SSE endpoint into the Airtable MCP Client node. Confirm the correct base and table IDs in each Airtable node. Expect faster content reviews, fewer manual edits, and clearer status checks for posts, campaigns, and tags.

• Chat trigger starts a conversation when a message arrives
• AI Agent powered by OpenAI to interpret requests and choose the right tool
• Simple memory window keeps recent context for multi step tasks
• MCP Server Trigger exposes Airtable actions as callable tools for the agent
• Airtable nodes for Get, Search, Create, Update, and Delete on a chosen base and table
• Search supports Airtable filterByFormula for precise lookups
• Record level operations by ID for safe updates and deletes
• Configurable OpenAI model selection with gpt 4o preset

• Reduce manual Airtable updates from 60 minutes to 5 minutes by using chat commands
• Streamline content ops by up to 70 percent with guided create and update actions
• Improve data accuracy by using structured CRUD steps instead of copy and paste
• Handle more requests at once as teammates can ask the agent in parallel
• Connect AI chat and Airtable so non technical users can manage records easily

1. Import the template into n8n: Create a new workflow in n8n > Click the three dots menu > Select 'Import from File' > Choose the downloaded JSON file.
2. You'll need accounts with Airtable and OpenAI. See the Tools Required section above for links to create accounts with these services.
3. In Airtable, create a Personal Access Token with read and write access to the target base. Copy the token.
4. In the n8n credentials manager, create a new Airtable Personal Access Token credential. Name it clearly, paste the token, and save.
5. Open each Airtable node (Get, Search, Create, Update, Delete) and pick your Airtable credential. Confirm the Base ID and Table ID match your Airtable base and table.
6. In your OpenAI account, create an API key. In the n8n credentials manager, create an OpenAI credential and paste the key.
7. Open the OpenAI Chat Model node and select your OpenAI credential. Keep the model as gpt 4o or choose another supported model.
8. Open the MCP Server Trigger node. Set a unique path value and save the workflow. Copy the webhook URL shown in the node; this is your MCP server endpoint.
9. Open the Airtable MCP Client node and paste the SSE endpoint URL from the MCP Server Trigger webhook. Save your changes.
10. Activate the workflow. In the n8n chat interface, send a message like Find draft posts to test the agent. Check that results appear and match Airtable.
11. If you see 401 errors, check the Airtable token scopes. If no messages arrive, verify the MCP path and webhook URL. For formula issues, test the filterByFormula directly in Airtable first.