Telegram + Google Gemini: daily chat digests
Your Telegram group moves fast. You look away for an hour, come back, and it’s 200 new messages, three decisions buried in side talk, and a “can someone resend that link?” thread you’ve seen before.
This is where Telegram digest automation pays for itself. Community managers feel it first, but project leads and small teams running client chats get hit too. You stop scrolling and start seeing what matters, daily.
This workflow grabs the last 24 hours of messages, asks Google Gemini to turn them into a structured digest, then posts it back into the same group. Below, you’ll see exactly what it solves, what you get, and how to set it up without getting technical.
How This Automation Works
The full n8n workflow, from trigger to final output:
n8n Workflow Template: Telegram + Google Gemini: daily chat digests
flowchart LR
subgraph sg0["Daily Schedule Flow"]
direction LR
n0["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/code.svg' width='40' height='40' /></div><br/>Filter and Shape Messages"]
n1@{ icon: "mdi:robot", form: "rounded", label: "Create Gemini Digest", pos: "b", h: 48 }
n2["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/telegram.svg' width='40' height='40' /></div><br/>Send Digest to Telegram"]
n3["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/code.svg' width='40' height='40' /></div><br/>Clean AI Summary"]
n4["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/httprequest.dark.svg' width='40' height='40' /></div><br/>Retrieve Telegram Updates"]
n5@{ icon: "mdi:play-circle", form: "rounded", label: "Daily Schedule Trigger", pos: "b", h: 48 }
n5 --> n4
n3 --> n2
n4 --> n0
n0 --> n1
n1 --> n3
end
%% Styling
classDef trigger fill:#e8f5e9,stroke:#388e3c,stroke-width:2px
classDef ai fill:#e3f2fd,stroke:#1976d2,stroke-width:2px
classDef aiModel fill:#e8eaf6,stroke:#3f51b5,stroke-width:2px
classDef decision fill:#fff8e1,stroke:#f9a825,stroke-width:2px
classDef database fill:#fce4ec,stroke:#c2185b,stroke-width:2px
classDef api fill:#fff3e0,stroke:#e65100,stroke-width:2px
classDef code fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px
classDef disabled stroke-dasharray: 5 5,opacity: 0.5
class n5 trigger
class n1 ai
class n4 api
class n0,n3 code
classDef customIcon fill:none,stroke:none
class n0,n2,n3,n4 customIcon
The Problem: Important Updates Get Lost in Chat Noise
Busy group chats don’t fail because people talk too much. They fail because the signal and the noise look identical when you’re catching up at speed. A key decision gets made at 10:12, someone drops a doc link at 10:18, then by 10:40 you’re deep in memes, quick questions, and “got it” replies. When you finally scroll back, you miss context, you misread tone, and you still don’t know what you’re supposed to do next. The cost is sneaky: repeated questions, slower work, and a lot of quiet frustration.
It adds up fast. Here’s where it usually breaks down in real teams.
- You end up spending about 20 minutes a day just trying to reconstruct what happened.
- Links get reposted (and reposted again), which clutters the chat and wastes everyone’s attention.
- Unanswered questions slip through because nobody can see them all in one place.
- People disengage from the group because catching up feels like homework.
The Solution: A Daily Telegram Digest Written by Gemini
This workflow runs quietly in the background and turns the last 24 hours of Telegram conversation into a clean “Daily Digest.” It starts on a daily schedule, pulls new messages from your Telegram group through your bot, then filters and shapes those messages into a transcript that’s easy for an AI model to understand. Next, Google Gemini 1.5 Flash generates a structured summary that calls out key topics, shared links, and open questions. After a quick cleanup pass (so formatting stays readable), the final digest gets posted back into the same Telegram group. No copy-paste. No “who has the summary?” scramble.
The flow is simple. A daily trigger fetches updates, your messages are cleaned up and organized, then Gemini produces the digest, and Telegram receives a polished post that the whole group can skim in a minute.
What You Get: Automation vs. Results
| What This Workflow Automates | Results You’ll Get |
|---|---|
|
|
Example: What This Looks Like
Say your client delivery group averages about 250 messages a day. Manually catching up is usually a 20-minute scroll (and another 10 minutes asking “wait, what did we decide?”). That’s roughly 30 minutes daily, five days a week. With this workflow, your “work” is basically zero: the schedule runs once a day, Gemini summarizes, and the digest lands in the chat. You skim it in about 2 minutes, then move on.
What You’ll Need
- n8n instance (try n8n Cloud free)
- Self-hosting option if you prefer (Hostinger works well)
- Telegram for the group chat source and delivery.
- Google Gemini to generate the daily digest summary.
- Telegram Bot Token (get it from @BotFather in Telegram)
- Google Gemini API key (get it from Google AI Studio)
Skill level: Beginner. You’ll paste tokens, select credentials in dropdowns, and toggle the workflow on.
Don’t want to set this up yourself? Talk to an automation expert (free 15-minute consultation).
How It Works
A daily schedule triggers the run. n8n starts this automation once every 24 hours, so your digest shows up reliably without anyone remembering to do it.
Telegram messages are pulled from the group. Using an HTTP request with your bot token, the workflow retrieves the latest updates from the groups your bot belongs to. If your bot can’t “see” messages, the workflow can’t summarize them, which is why privacy settings matter.
The chat is cleaned and shaped for summarization. A filtering step removes the junk (like empty messages) and organizes the transcript so the AI model receives something coherent instead of a messy log dump. Honestly, this is what makes the digest readable.
Gemini generates the digest and Telegram receives it. Google Gemini produces a structured daily summary (key topics, links, open questions), then n8n applies a final cleanup pass and sends a formatted message back into the same Telegram group.
You can easily modify the digest format to match your team’s style based on your needs. See the full implementation guide below for customization options.
Step-by-Step Implementation Guide
Step 1: Configure the Schedule Trigger
Set up the workflow to run on a daily schedule using the trigger node.
- Add the Daily Schedule Trigger node.
- Open Daily Schedule Trigger and configure the Rule interval to your preferred daily time (the current rule is empty and needs a schedule).
- Confirm Daily Schedule Trigger connects to Retrieve Telegram Updates.
Step 2: Connect Telegram Updates Retrieval
Pull the latest Telegram updates so the workflow can extract the last 24 hours of messages.
- Add the Retrieve Telegram Updates node.
- Set URL to
https://api.telegram.org/bot[CONFIGURE_YOUR_TOKEN]/getUpdates. - Keep the Timeout option at
3000unless you need a longer wait time. - Connect Daily Schedule Trigger → Retrieve Telegram Updates.
[CONFIGURE_YOUR_TOKEN] with your Telegram Bot API token, or the request will fail.Step 3: Set Up Message Processing and AI Digest
Filter, format, and summarize the chat transcript using code and Gemini.
- Add the Filter and Shape Messages node and paste the provided JavaScript Code exactly as shown to build the
transcript. - Connect Retrieve Telegram Updates → Filter and Shape Messages.
- Add the Create Gemini Digest node and set Model to
models/gemini-1.5-flash-latest. - In Create Gemini Digest → Messages, set the content to the full prompt and include the transcript expression:
{{ $('Filter and Shape Messages').item.json.transcript }}. - Add the Clean AI Summary node and paste the provided JavaScript Code to escape MarkdownV2 for Telegram.
- Ensure Clean AI Summary references Gemini output using
$('Create Gemini Digest').first().json.content.parts[0].text.
Step 4: Configure Telegram Delivery
Send the cleaned digest to your Telegram chat.
- Add the Send Digest to Telegram node.
- Set Text to
{{ $json.text }}. - Set Chat ID to
[YOUR_ID](replace with your actual chat or channel ID). - Set Parse Mode to
MarkdownV2in Additional Fields. - Connect Clean AI Summary → Send Digest to Telegram.
Step 5: Test and Activate Your Workflow
Run a manual test to confirm the digest is generated and delivered, then activate the schedule.
- Click Execute Workflow to run Daily Schedule Trigger manually.
- Verify Retrieve Telegram Updates returns updates and Filter and Shape Messages produces a
transcript. - Confirm Create Gemini Digest generates a response and Clean AI Summary outputs
text. - Check Telegram for the delivered digest from Send Digest to Telegram formatted in MarkdownV2.
- Toggle the workflow to Active to enable daily execution.
Common Gotchas
- Telegram credentials and bot settings are usually the culprit. If the workflow retrieves nothing, check @BotFather first and confirm Group Privacy Mode is turned OFF.
- If you’re using Wait nodes or external rendering, processing times vary. Bump up the wait duration if downstream nodes fail on empty responses.
- Default prompts in AI nodes are generic. Add your brand voice early or you’ll be editing outputs forever.
Frequently Asked Questions
About 10 minutes if your bot and API key are ready.
No. You’ll paste credentials and choose them in n8n node dropdowns.
Yes. n8n has a free self-hosted option and a free trial on n8n Cloud. Cloud plans start at $20/month for higher volume. You’ll also need to factor in Google Gemini API usage (often just a few dollars a month for typical digests).
Two options: n8n Cloud (managed, easiest setup) or self-hosting on a VPS. For self-hosting, Hostinger VPS is affordable and handles n8n well. Self-hosting gives you unlimited executions but requires basic server management.
Yes, and it’s the most common tweak. Change the prompt and structure in the “Create Gemini Digest” node so it outputs your preferred sections (for example: decisions first, then action items, then links). You can also adjust the cleanup step after Gemini runs to enforce formatting. Some teams add a short “high priority” line at the top so the first sentence tells you if you need to act today.
Most of the time it’s the bot setup, not n8n. Make sure the bot is actually in the group you want to summarize, then check @BotFather and confirm Group Privacy Mode is OFF so it can read messages. If you pasted the bot token into the HTTP request URL, re-check for a missing character. Finally, Telegram can rate limit noisy bots, so avoid fetching too frequently while testing.
A lot, but huge chats may need trimming so the transcript fits within model limits. In practice, most teams summarize one active group per day without issues, and if your group is extremely busy you can summarize more often (morning + afternoon) or only include messages that contain links, questions, or keywords.
Often, yes, because n8n is better for multi-step shaping and cleanup before and after the AI summary. You can run it on a schedule, filter and format messages the way you want, and self-host if you need unlimited executions. Zapier or Make can work for simpler “send transcript to AI, post result” setups, but they get awkward once you start cleaning transcripts and enforcing a digest template. If you’re torn, Talk to an automation expert and we’ll map it to your exact chat volume and needs.
Once this is live, the group gets clarity every day without anyone playing catch-up detective. Set it up once, then let the digest do the heavy lifting.
Need Help Setting This Up?
Our automation experts can build and customize this workflow for your specific needs. Free 15-minute consultation—no commitment required.
