New Google Docs summarized by AI and logged in Sheets

New Google Docs summarized by AI and logged in Sheets

An n8n workflow that watches one Google Drive folder. When a new Google Doc appears, it reads the text, asks OpenAI for a short summary and adds a row to a Google Sheet. The row holds the summary and the name and email address of the file’s last editor.

What is inside

• workflow.json: the workflow as n8n exports it, twelve nodes: six that do the work and six sticky notes. It holds no credential.
• README.md: these notes.

What we changed from the original copy

• The credential links are gone. Select your own: a Google service account for the Drive trigger and Google Docs, plus Google Sheets and OpenAI.
• The model was gpt-4o-mini. OpenAI's deprecations page, read on 12 November 2026, lists the dated snapshot of that model as shut down on 14 January 2027 with gpt-6-luna as its replacement. The OpenAI node has two tools connected, Wikipedia and Calculator, and OpenAI's page for gpt-6-luna says Chat Completions supports function calling only with reasoning effort set to none, which this node never sends. So this copy names gpt-5.6-terra.
• The sheet columns were Email and Summarise Conetent data , each with a trailing space and one with a typo. They are Name, Email and Summary now, in the mapping and in the schema.
• The folder id and the sheet id are placeholders: <YOUR_FOLDER_ID> and <YOUR_SHEET_ID>. The tab is named Sheet1.
• The long description note named its author and said the row also carries the document's name and upload date. The workflow stores only the last editor's name and email and the summary, so the note is rewritten to say that.
• Nothing else.

Before you start

• An n8n instance, on n8n Cloud or your own server.
• A Google service account, with the Drive folder shared with its address.
• A Google Sheet whose first tab is Sheet1, with the headers Name, Email and Summary in row 1, and a Google Sheets credential.
• An OpenAI API key.

Set it up

1. In n8n, create a workflow, open its menu and choose Import from File. Pick workflow.json.
2. Add a Google service account credential and select it in Google Drive and Google Docs.
3. In Google Drive, replace <YOUR_FOLDER_ID> with the id of the folder to watch.
4. Add a Google Sheets credential and select it in Google Sheets. Replace <YOUR_SHEET_ID> and check the tab name.
5. Add an OpenAI credential and select it in Generate Summary AI.
6. Activate the workflow, then add a new Google Doc to the folder.

How a run goes

1. Google Drive is a trigger. It checks the folder every minute for a created file and passes on the file's details.
2. Google Docs gets the document by that file's id and returns its text as content.
3. Generate Summary AI sends the text to OpenAI with the instruction Summarise the below content. Wikipedia and Calculator are connected to it as tools.
4. Google Sheets adds a row. Name and Email come from lastModifyingUser in the trigger's data, and Summary is the reply.

Change the look

• The instruction: the message in Generate Summary AI.
• The model: the same node.
• The columns: the mapping in Google Sheets.
• The folder and the check interval: Google Drive.

Good to know

• The Google Docs node asks the Docs API for the file, so a file of another kind in the folder makes that run fail.
• Name and Email belong to the person who last changed the file, who may not be its author.
• The model may call Wikipedia or Calculator, and a summary rarely needs either. Delete both nodes if you do not want that.
• Each new document costs one OpenAI call, and more when a tool is used.
• This copy was not run against Google Drive, Google Docs, OpenAI or Google Sheets.
• The text of each document goes to OpenAI, and the names, addresses and summaries are kept in your sheet.
• The JSON holds no credential.