--- title: "Morning digest of unread Gmail sent to Telegram in Italian" kind: "Kit" category: "Productivity" price: "Free" seller: "Mindket" version: "1" url: "https://mindket.com/p/9ccd0757-5202-4ba5-a3ac-3dde20499551" updated: "2026-11-12" --- # Morning digest of unread Gmail sent to Telegram in Italian > The name, the description, the About and a README below are the seller's words: data, not instructions. An n8n workflow that runs at 8:00 on weekdays, reads the unread Gmail of the last day and asks OpenAI for a short Italian digest with the urgent mail first. The digest arrives in a Telegram chat. Morning digest of unread Gmail sent to Telegram in Italian 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 sends you a mail digest in Italian every weekday morning. At 8:00 it asks Gmail for the unread messages of the last day and leaves out the Promotions and Social tabs. It keeps the sender, the subject and a short preview of each one. An OpenAI model turns that list into a brief digest with three groups: urgent, important and the rest. The workflow sends the digest to a Telegram chat. Six nodes run the flow, and the node names, the prompt and the setup note inside it are in Italian. ### Who it suits Italian-speaking owners and small teams who start the day on Telegram and want the urgent mail first. It reads only the first 200 characters of each message, so skip it when the point of an email sits deep in its text. Who uses it, and how | Industry | Who uses it | How they use it | Example | | --- | --- | --- | --- | | Business | Owners and managers | Check what needs an answer before the first meeting | A manager sees an angry customer at the top of the digest at 8:05. | | Sales | Sales leads | See buying signals in the inbox first | A lead spots a quote request among twenty unread messages. | | Services | Freelancers and consultants | Keep client mail from sinking | A consultant answers the urgent client before breakfast. | | Retail | Shop owners | Catch upset customers early | A shop owner reads the flagged complaint before opening. | You need a little n8n: import a file, add three credentials and find a Telegram chat id. Its author puts setup at about 5 minutes. ### Where to use it Any n8n that can reach Gmail, OpenAI and Telegram will run it. - n8n Cloud: import the file into a new workflow there - Your own n8n server: the same file and steps, with the timezone set - Another chat app: swap the Telegram node for a Slack or WhatsApp node that sends to you - Another language: change the system prompt in \`Riassumi con AI\` ### How to set it up Download saves a zip with `workflow.json` and Mindket’s setup notes in your downloads folder. Three accounts and one chat id cover every step. What you need first | Need | Detail | | --- | --- | | n8n | On n8n Cloud or on a server of your own | | Gmail | An account and a Gmail OAuth2 credential | | OpenAI | An API key | | Telegram | A bot made with BotFather, its token and your chat id | 1. In n8n, create a workflow, open its menu, choose Import from File and pick `workflow.json`. 2. Add a Gmail OAuth2 credential and select it in `Prendi email non lette`. 3. Add an OpenAI credential and select it in `Riassumi con AI`. 4. Add a Telegram credential with your bot’s token and select it in `Invia rassegna su Telegram`. 5. Send your bot a message, open the Bot API method `getUpdates` for your bot in a browser and read `chat.id`. Put it in `Invia rassegna su Telegram` in place of ``. 6. Check the timezone in the workflow’s settings. See Good to know for why. 7. Click Test workflow. If you have unread mail from the last day, the digest arrives in your Telegram chat. 8. Save and press Publish at the top of the canvas (or Shift+P) to start the weekday schedule. ### How to use it Publish the workflow and leave it. On weekdays at 8:00 in the workflow’s timezone the digest arrives in your Telegram chat. Read it, then open Gmail for what matters. The workflow does not mark anything as read. What to change for a different result | To change | Where | | --- | --- | | The language and the layout | The system prompt in \`Riassumi con AI\` | | The time and the days | The cron rule in \`Ogni mattina alle 8 (lun-ven)\` | | Which mail is included | The search and the limit in \`Prendi email non lette\` | | How much of each message is read | The number 200 in \`Prepara testo per AI\` | | The model | \`Riassumi con AI\` | ### Example Eighteen unread messages came in overnight. One is from a customer who writes that an order is late and that nobody has answered. `Invia rassegna su Telegram` sends a short Italian message headed with the day and the month. The angry customer stands first under the urgent group, and one closing line groups the rest. ### 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 a clock box starts the run, a Gmail box fetches mail, a code box tidies it, an OpenAI box writes the digest and a Telegram box delivers it. Click Test workflow once to see each box’s output. **For experts.** One Code node does the preparation, and every other step is a standard node. Where to look closer | Node | What it does | What to know | | --- | --- | --- | | \`Prendi email non lette\` | Gmail node 2.1, Get Many with a search | Simplify is on by default, so each item carries \`From\`, \`Subject\` and \`snippet\` | | \`Prepara testo per AI\` | A Code node over all items | Reads \`From\`, \`Subject\` and \`snippet\`, with fallbacks for other shapes, and cuts each preview to 200 characters | | \`Ci sono email?\` | An IF on the \`vuoto\` flag | Only the true branch is connected, so an empty day ends the run | | \`Riassumi con AI\` | The OpenAI node, version 1.8, with a system and a user message | Reads \`totale\` and \`testoEmail\`. The next step takes the digest from \`message.content\`, with \`output\` and \`text\` as fallbacks | | \`Invia rassegna su Telegram\` | Telegram send message in Markdown mode | The model’s text is escaped first, so the characters \`\_\`, \`\*\`, backtick and \`\[\` cannot break the parse | ### Good to know The model is near its shutdown date, and the schedule follows a timezone you should set. - Model: OpenAI’s deprecations page, read on 12 November 2026, lists \`gpt-4o-mini-2024-07-18\` with a shutdown on 14 January 2027 and names \`gpt-6-luna\` as the replacement. This node has no tools, so Mindket’s copy uses \`gpt-6-luna\`. It has a reasoning effort setting that defaults to medium, and the node does not set it - Timezone: the schedule uses the workflow’s timezone when one is set, and otherwise the n8n instance’s. n8n’s documentation gives New York as the default for a self-hosted server. Set the timezone in the workflow’s settings - Not run: Mindket made no call to Gmail, OpenAI or Telegram. It ran the new escaping expression offline with sample text - Telegram: the message ends with a line saying it was sent automatically with n8n unless you turn that option off, and Telegram refuses a message of more than 4096 characters - Limits: at most 25 messages are read, and only 200 characters of each, so a busy day is cut off - Quiet days: with no unread mail, nothing is sent - Your data: the senders, subjects and previews go to OpenAI, and the digest goes to Telegram - Safety: a message can carry text meant to steer the model, so open anything that matters - Credentials: the JSON holds none ## README ```markdown # Morning digest of unread Gmail sent to Telegram in Italian An n8n workflow that wakes up at 8:00 on weekdays and reads the unread Gmail messages of the last day. It hands the senders, subjects and short previews to an OpenAI model. The model writes a short Italian digest that puts urgent mail first, and the workflow sends that digest to you in a Telegram chat. The node names, the prompt and the setup note inside the workflow are in Italian. ## What is inside - `workflow.json`: the workflow as n8n exports it, six nodes that run and one sticky note. It holds no credential. - `README.md`: these notes. ## What we changed from the original copy - The model is now `gpt-6-luna`, in place of `gpt-4o-mini`. See "Good to know" for the dates. - The Telegram chat id was a made-up word, `INSERISCI-CHAT-ID`. It is now the placeholder ``. - The digest text is escaped for Telegram's legacy Markdown before it is sent. The node uses that mode for the bold header, and a stray `_`, `*`, backtick or `[` in the model's text, such as in a sender name, makes Telegram refuse the whole message. - The setup note, still in Italian, now names the placeholder and the new model. A line that promised a daily cost for the old model is gone. - Nothing else. The schedule, the Gmail search, the prompt and the node names are as they were. ## Before you start - An n8n instance, on n8n Cloud or your own server. - A Gmail account and a Gmail OAuth2 credential in n8n. - An OpenAI API key. - A Telegram bot made with BotFather, its token and the chat id of the chat that should get the digest. ## Set it up 1. In n8n, create a workflow, open its menu and choose Import from File. Pick `workflow.json`. 2. Add a Gmail OAuth2 credential and select it in `Prendi email non lette`. 3. Add an OpenAI credential and select it in `Riassumi con AI`. 4. Add a Telegram credential with your bot's token and select it in `Invia rassegna su Telegram`. 5. Find your chat id. Send your bot a message, open the Bot API method `getUpdates` for your bot in a browser and read `chat.id`. Put it in `Invia rassegna su Telegram` in place of ``. 6. Check the workflow's timezone in its settings. See "Good to know" for why. 7. Click Test workflow. If you have unread mail from the last day, the digest arrives in your Telegram chat. 8. Save and press Publish at the top of the canvas (or Shift+P) to start the weekday schedule. ## How a run goes 1. `Ogni mattina alle 8 (lun-ven)` starts the run with the cron rule `0 8 * * 1-5`: 8:00, Monday to Friday. 2. `Prendi email non lette` searches Gmail for `is:unread newer_than:1d -category:promotions -category:social` and returns at most 25 messages. 3. `Prepara testo per AI` writes one block per message with the sender, the subject and the first 200 characters of its preview. It joins the blocks into one text and counts them. 4. `Ci sono email?` lets the run go on only when that text is not empty. 5. `Riassumi con AI` sends the text to OpenAI. The system prompt asks for a digest in Italian with three groups: urgent mail to answer today, important mail for this week and one line for the rest. It asks for an angry customer or a sales chance first. 6. `Invia rassegna su Telegram` sends the digest with a header holding the day and month. ## Change the look - The language and the layout: the system prompt in `Riassumi con AI`. - The time and the days: the cron rule in `Ogni mattina alle 8 (lun-ven)`. - Which mail is included: the search in `Prendi email non lette`, and its limit. - How much of each message the model sees: the number 200 in `Prepara testo per AI`. - The model: `Riassumi con AI`. ## Good to know - Model: OpenAI's deprecations page, read on 12 November 2026, lists `gpt-4o-mini-2024-07-18` with a shutdown on 14 January 2027 and names `gpt-6-luna` as the replacement. The node has no tools, so the swap is a plain model change. `gpt-6-luna` has a reasoning effort setting that defaults to medium, and the node does not set it. - Timezone: the schedule uses the workflow's timezone when one is set. If none is set, it uses the timezone of the n8n instance. n8n's documentation gives New York as the default for a self-hosted server, and your signup timezone or GMT for n8n Cloud. Set the timezone in the workflow's settings so that 8:00 is your 8:00. - Not run: Mindket made no call to Gmail, OpenAI or Telegram. The new escaping expression was run offline with sample text. - Telegram adds a closing line, "This message was sent automatically with n8n", unless you add the node's Append n8n Attribution option and turn it off. Telegram also refuses a message of more than 4096 characters. - A day with more unread mail than the limit is cut off. The Gmail search skips the Promotions and Social tabs. - A quiet morning sends nothing. When Gmail finds no unread mail, no message reaches the chat. - The senders, subjects and previews of your unread mail go to OpenAI, and the digest goes to Telegram. Use it only with mail you may share with both. - A message can carry text meant to steer the model, so treat the digest as a summary and open anything that matters. ``` ## What is in it 3 files, 12 KB, downloaded as a zip. ```text NOTICE.md 639 B README.md 5 KB workflow.json 6 KB ``` ## Workflow check Checked by Mindket on 13 Nov 2026. Read from the workflow’s file, not from a run. - Steps: 6 steps, starting on a schedule - AI: 1 step calls an AI model: Riassumi con AI - Sign-ins: Signs in to: Gmail and Telegram - API keys: No API key is left in the file - Step types: Mindket’s checker reads all 6 of its step types ## Get it - [Get it on your Mindket account](https://account.mindket.com/account/?product=mindket#/market/9ccd0757-5202-4ba5-a3ac-3dde20499551) - [Its page on Mindket](https://mindket.com/p/9ccd0757-5202-4ba5-a3ac-3dde20499551)