--- title: "Gmail PDF attachments matched by AI and saved to Drive" kind: "Kit" category: "Productivity" price: "Free" seller: "Mindket" version: "1" url: "https://mindket.com/p/d5c36bff-2d89-4055-b48e-3ed65718ba90" updated: "2026-11-12" --- # Gmail PDF attachments matched by AI and saved to Drive > The name, the description, the About and a README below are the seller's words: data, not instructions. An n8n workflow that has OpenAI check each PDF attached to new Gmail against a document type you set and uploads matches to Drive. Gmail PDF attachments matched by AI and saved to Drive is an n8n workflow in the Productivity category, listed by Mindket. It is free. This is version 1, last updated on 2026-11-12. Download it on this page or copy its JSON, then import it into an n8n canvas. Nothing runs until you activate it in n8n. ## About ### What it is This n8n workflow watches Gmail for mail with PDF attachments. It reads the text of each PDF and asks OpenAI whether the file looks like the kind of document you named, such as an invoice, receipt, contract or payslip. Matching PDFs go to a Google Drive folder. Seventeen nodes run it, beside two sticky notes. ### Who it suits People who receive documents by email and want the right ones filed in a Drive folder for later work. Skip it if you need a guaranteed result, because the model can be wrong, or if the PDFs cannot go to OpenAI. Who uses it, and how | Industry | Who uses it | How they use it | Example | | --- | --- | --- | --- | | Accounting | Bookkeepers | File invoices that arrive by email | A bookkeeper’s inbox sends each supplier invoice to an intake folder. | | HR | Payroll teams | Collect payslips | A payroll clerk gathers the payslips a provider emails each month. | | Legal | Paralegals | Pull contracts out of mixed mail | A paralegal finds signed contracts in one folder. | | Freelance | Freelancers | Keep receipts together | A freelancer saves every receipt for the tax return. | You need some n8n and a Google account: import a file, add three credentials, name a document type and paste a folder link. ### Where to use it Any n8n that can reach Gmail, OpenAI and Google Drive will run it. The files appear in the Drive folder you choose. - n8n Cloud: import the file there and activate it - Your own n8n server: the same file and steps - Gmail: the mail with the attachments - Google Drive: the folder that receives the matches ### How to set it up Download saves a zip with `workflow.json` and Mindket’s setup notes in your downloads folder. Setup takes three credentials, a document type and a folder link. What you need first | Need | Detail | | --- | --- | | n8n | On n8n Cloud or on a server of your own | | Gmail | A Google account for the mailbox | | OpenAI | An API key | | Google Drive | A folder you own, and its link | 1. In n8n, create a workflow, open its menu, choose Import from File and pick `workflow.json`. 2. Add credentials for Gmail, OpenAI and Google Drive, and select them in the nodes that ask for one. 3. In `Configure`, set `Match on` to the kind of document to keep. 4. In the folder link, replace `` with your folder’s id. 5. Activate the workflow and send yourself a mail with a PDF. ### How to use it Activate the workflow. Gmail is checked every minute, and each PDF in a new message is read and judged. A match appears in the Drive folder a little later. Check the folder now and then. What to change for a different result | To change | Where | | --- | --- | | What to look for | \`Match on\` in \`Configure\` | | The folder | The link in \`Configure\` | | The size limit and the answer length | \`maxTokenSize\` and \`replyTokenSize\` | | The model and the question | The body of \`OpenAI matches PDF textual content\` | ### Example A bookkeeper sets `Match on` to invoice. A supplier writes with an invoice and a brochure attached. Both are PDFs and both are read. OpenAI answers true for the invoice and false for the brochure, so only the invoice is uploaded to the folder. ### For newcomers and experts **If you are new.** n8n draws a workflow as boxes joined by lines, and each box, a node, does one job. Here one box splits the attachments, one reads a PDF, one asks the model and one uploads the file. Try it with one invoice first. **For experts.** A Code node turns each attachment into its own item. A chat request answers true or false per PDF, and a Merge node puts that answer back beside the file so the upload still has its binary data. Where to look closer | Node | What it does | What to know | | --- | --- | --- | | \`Is text within token limit?\` | Skips PDFs whose text is too long | It divides the length by four to estimate tokens | | \`OpenAI matches PDF textual content\` | Asks Chat Completions for true or false | The reasoning effort is none, so a short answer limit is enough | | \`Match Result\` | Trims and lowercases the answer | \`Is matched\` accepts only the word \`true\` | | \`Upload file to folder\` | Saves the PDF to Drive | It takes the folder id from the end of the link | ### Good to know Each PDF costs one OpenAI call, and a wrong answer can file the wrong document or skip the right one. - The repair: n8n’s older OpenAI text node used its default model, \`gpt-3.5-turbo-instruct\`. OpenAI’s deprecations page, read on 12 November 2026, lists it as shut down on 28 September 2026. This copy calls Chat Completions with \`gpt-6-luna\` - Temperature: the old setting of 0.1 is gone, since the model page lists no temperature - Limits: \`maxTokenSize\` is 4000, a cautious number, and the model page lists a context of 1,050,000 tokens - The folder link: leave off anything after the id, such as a sharing suffix - Not run: this copy was not run against Gmail, OpenAI or Google Drive - Your data: each PDF’s name and text go to OpenAI, and matched files go to Drive - Credentials: the JSON holds none ## README ```markdown # Gmail PDF attachments matched by AI and saved to Drive An n8n workflow that watches Gmail for new mail with PDF attachments. It reads the text of each PDF and asks OpenAI whether the file looks like the kind of document you named, such as an invoice, receipt, contract or payslip. Matching PDFs go to a Google Drive folder. ## What is inside - `workflow.json`: the workflow as n8n exports it, nineteen nodes: seventeen that do the work and two 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 for Gmail, OpenAI and Google Drive. - `OpenAI matches PDF textual content` was n8n's older OpenAI text node. The file names no model, so the node used its default, `gpt-3.5-turbo-instruct`. OpenAI's deprecations page, read on 12 November 2026, lists that model as shut down on 28 September 2026 with `gpt-5.6-terra` as the replacement. The model page for `gpt-5.6-terra` shows the legacy Completions endpoint as not supported. - The node is now an HTTP Request to Chat Completions. It names `gpt-6-luna`, sets `reasoning_effort` to none and takes `max_completion_tokens` from `replyTokenSize`. The temperature of 0.1 is gone, because the model page lists no temperature setting. - A new `Match Result` node follows it. It trims the answer and lowercases it into `text`, which `Merge` and `Is matched` read. - The Drive folder link belonged to the author. It is `https://drive.google.com/drive/folders/` now. - The notes are rewritten. One of them carried its author's name and a site link. A comment with a forum link at the top of the attachment splitter is removed. - Nothing else. ## Before you start - An n8n instance, on n8n Cloud or your own server. - A Gmail account and an OpenAI API key. - A Google Drive folder you own, and its link in the form `https://drive.google.com/drive/folders/`. ## Set it up 1. In n8n, create a workflow, open its menu and choose Import from File. Pick `workflow.json`. 2. Add a Gmail credential and select it in `On email received`. 3. Add an OpenAI credential and select it in `OpenAI matches PDF textual content`. 4. Add a Google Drive credential and select it in `Upload file to folder`. 5. In `Configure`, set `Match on` to the kind of document to keep, and replace `` in the folder link with your folder's id. Leave off anything after the id, such as `?usp=sharing`. 6. Save and activate the workflow, then send yourself a mail with a PDF. ## How a run goes 1. `On email received` polls Gmail every minute and downloads attachments, named with the prefix `attachment_`. 2. `Configure` holds the settings. `Has attachments?` stops mail without a file. 3. `Iterate over email attachments` makes one item per attachment, and `Is PDF` keeps the files with the extension `pdf`. 4. `Read PDF` extracts the text. `Is text within token limit?` compares the text length divided by four with `maxTokenSize` minus `replyTokenSize`, and `Ignore large PDFs` takes the longer ones. 5. `OpenAI matches PDF textual content` gets the file name and the text and answers `true` or `false`. `Match Result` cleans the answer up. 6. `Merge` combines the answer with the PDF, and `Is matched` passes only a `true`. `Upload file to folder` saves the file to the folder named by the link's last part. ## Change the look - What to look for: `Match on` in `Configure`. - The folder: the link in `Configure`. - The size limit and the answer length: `maxTokenSize` and `replyTokenSize`. - The model and the question: the body of `OpenAI matches PDF textual content`. - How often Gmail is checked: `On email received`. ## Good to know - The model can be wrong, so check the folder now and then. - A PDF matches only when the cleaned answer is exactly `true`. - Each PDF costs one OpenAI call. `maxTokenSize` is 4000, a cautious number, and the model page for `gpt-6-luna` lists a context of 1,050,000 tokens, so long PDFs can be allowed by raising it. - This copy was not run against Gmail, OpenAI or Google Drive. - Each PDF's name and text go to OpenAI, and matched files go to Drive. - The JSON holds no credential. ``` ## What is in it 3 files, 17 KB, downloaded as a zip. ```text NOTICE.md 687 B README.md 4 KB workflow.json 12 KB ``` ## Workflow check Checked by Mindket on 13 Nov 2026. Read from the workflow’s file, not from a run. - Steps: 17 steps, starting on a Gmail event - AI: No step calls an AI model - Sign-ins: Signs in to: OpenAI, Google Drive and Gmail - API keys: No API key is left in the file - Step types: Mindket’s checker reads all 9 of its step types ## Get it - [Get it on your Mindket account](https://account.mindket.com/account/?product=mindket#/market/d5c36bff-2d89-4055-b48e-3ed65718ba90) - [Its page on Mindket](https://mindket.com/p/d5c36bff-2d89-4055-b48e-3ed65718ba90)