January 22, 2026

Telegram + Google Sheets: LinkedIn drafts on demand

Lisa Granqvist Partner Workflow Automation Expert

Get a free AI assessment → ⬇️ Use template

You find a great article, think “I should post about this,” then it disappears into 14 open tabs. Later, you’re staring at a blank LinkedIn draft, trying to remember what the piece even said.

This Telegram Sheets drafts automation hits marketers hardest, but founders and solo consultants feel it too. You’ll turn random links into a searchable library, then request a LinkedIn-ready draft on demand without reopening the article.

Below you’ll see exactly how the workflow captures links, summarizes them with AI, saves them to Google Sheets, and sends you a clean post draft when you type one simple command.

How This Automation Works

The full n8n workflow, from trigger to final output:

n8n Workflow Template: Telegram + Google Sheets: LinkedIn drafts on demand

Click to explore

flowchart LR

    subgraph sg0["Telegram Flow"]
        direction LR
        n0@{ icon: "mdi:swap-vertical", form: "rounded", label: "Edit Fields", pos: "b", h: 48 }
        n1@{ icon: "mdi:robot", form: "rounded", label: "Content collector", 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/>Telegram Trigger"]
        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/telegram.svg' width='40' height='40' /></div><br/>Send a text message"]
        n4@{ icon: "mdi:swap-horizontal", form: "rounded", label: "If generate", pos: "b", h: 48 }
        n5["<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 Code"]
        n6@{ icon: "mdi:swap-horizontal", form: "rounded", label: "article or generate?", pos: "b", h: 48 }
        n7@{ icon: "mdi:swap-vertical", form: "rounded", label: "Set Field", pos: "b", h: 48 }
        n8@{ icon: "mdi:database", form: "rounded", label: "Append row in sheet", pos: "b", h: 48 }
        n9@{ icon: "mdi:database", form: "rounded", label: "Get row(s) in sheet in Googl..", pos: "b", h: 48 }
        n10["<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/>The post"]
        n11@{ icon: "mdi:robot", form: "rounded", label: "Generate LinkedIn Post", pos: "b", h: 48 }
        n12@{ icon: "mdi:brain", form: "rounded", label: "OpenAI Chat Model", pos: "b", h: 48 }
        n13@{ icon: "mdi:brain", form: "rounded", label: "OpenAI Chat Model1", pos: "b", h: 48 }
        n7 --> n1
        n5 --> n8
        n0 --> n5
        n4 --> n11
        n2 --> n6
        n1 --> n0
        n12 -.-> n1
        n13 -.-> n11
        n8 --> n3
        n6 --> n7
        n6 --> n4
        n11 --> n10
        n9 -.-> n11
    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 n2 trigger
    class n1,n11 ai
    class n12,n13 aiModel
    class n4,n6 decision
    class n8,n9 database
    class n5 code
    classDef customIcon fill:none,stroke:none
    class n2,n3,n5,n10 customIcon

The Problem: Good articles don’t become good posts

Saving links isn’t the hard part. The hard part is turning those links into something you can actually publish. You skim an article, you think you’ll remember the hook, then client work happens and it’s gone. When you finally sit down to write, you’re re-reading the same content, re-highlighting the same points, and still second-guessing the angle. Worse, your “swipe file” lives across bookmarks, DMs, screenshots, and notes. The result is inconsistent posting, and a weird feeling that you’re always behind.

It adds up fast. Here’s where it breaks down in real life.

You lose at least 15 minutes per article re-finding and re-reading key parts.
Your ideas aren’t searchable, so “I know I saved something on this…” becomes a time sink.
Draft quality swings because you’re writing from memory instead of structured notes.
Posting slows down since writing starts from zero every time.

The Solution: Telegram in, LinkedIn drafts out

This workflow gives you a simple habit: when you see an article worth sharing, you just send the link to your Telegram bot. n8n grabs the URL, fetches the content, and hands it to an AI “insight gatherer” that summarizes the piece and pulls out the main arguments in a structured way. Then it maps those outputs into consistent fields and appends a new row in Google Sheets, so you’re building a real content library over time. When you’re ready to post, you type “generate” in Telegram. The workflow retrieves rows from your sheet and uses AI again to compose a LinkedIn-ready post draft, then sends it back to Telegram for quick editing and copy/paste.

The workflow starts with a Telegram message. If it’s a link, it gets analyzed and saved to Google Sheets with a clean summary and commentary. If it’s “generate,” the workflow pulls from the sheet and delivers a formatted LinkedIn draft back to you in Telegram.

What You Get: Automation vs. Results

What This Workflow Automates

Results You’ll Get

It detects whether your Telegram message is a link or the “generate” command.
It fetches article content (including PDFs when needed) and standardizes it for AI.
It summarizes the article and extracts key points into consistent fields.
It appends each new item to Google Sheets and retrieves saved rows on demand.

Most teams get about 3–5 hours back every week from less re-reading and rewriting.
Your best ideas become searchable in one Google Sheet instead of scattered links.
Drafts are more consistent because every post starts with structured insights.
You can post more often without “writing day” taking over your calendar.
You build a reusable content engine that gets better the more you use it.

Example: What This Looks Like

Say you save 10 articles a week. Manually, you might spend about 20 minutes per article to re-open it, pull quotes, and write a rough draft, which is roughly 3 hours. With this workflow, you spend maybe 1 minute sending each link to Telegram, and later you type “generate” and wait a minute for the draft to return. That’s about 15 minutes of effort for the same 10-post pipeline, with the research already captured in your sheet.

What You’ll Need

n8n instance (try n8n Cloud free)
Self-hosting option if you prefer (Hostinger works well)
Telegram for collecting links and requesting drafts
Google Sheets to store summaries and insights
OpenAI API key (or another LLM provider) (get it from your AI provider dashboard)

Skill level: Beginner. You’ll connect accounts, paste API keys, and match a few Google Sheet columns.

Don’t want to set this up yourself? Talk to an automation expert (free 15-minute consultation).

How It Works

A Telegram message kicks it off. The Telegram bot is your single input. Drop an article URL when inspiration hits, or type “generate” when you want a draft back.

The workflow decides what you meant. An “If” check routes the message: links go to analysis, while the generate command goes to drafting. Small detail, big usability win because you don’t need separate commands for “save” versus “write.”

AI turns raw content into usable notes. The Insight Gatherer agent plus the OpenAI chat model analyzes the page (or PDF), summarizes it, and pulls out key arguments and context. Then “Set” and normalization steps shape the data so every row in Sheets follows the same structure.

Google Sheets becomes your content library. New links are appended as rows with fields like date, title, link, summary, and insights commentary. When you request a post, the workflow retrieves rows and sends them into a second AI agent that composes a LinkedIn-style draft, which is delivered right back to Telegram.

You can easily modify the selection logic to pull the latest item, a random row, or a specific topic based on your needs. See the full implementation guide below for customization options.

Step-by-Step Implementation Guide

Step 1: Configure the Telegram Trigger

Set up the workflow entry point so Telegram messages can kick off the automation.

Add and configure Telegram Intake Trigger to listen for updates of type message.
Credential Required: Connect your telegramApi credentials in Telegram Intake Trigger.
Verify the trigger is connected to Detect Link or Command so the workflow can branch based on message content.

Tip: Keep Flowpast Branding as a visual label only—it doesn’t affect execution.

Step 2: Connect Google Sheets

Link your spreadsheet to store and retrieve content summaries and insights.

Open Append Sheet Row and select the target Document and Sheet in documentId and sheetName.
Set operation to append in Append Sheet Row.
Credential Required: Connect your googleSheetsOAuth2Api credentials in Append Sheet Row.
Configure Retrieve Sheet Rows with the same documentId and sheetName values so the AI can access existing records.
Credential Required: Connect your googleSheetsOAuth2Api credentials (this is an AI tool for Compose LinkedIn Post—ensure credentials are added at the parent agent configuration, not the tool node).

⚠️ Common Pitfall: Leaving documentId or sheetName empty will cause the append and tool calls to fail.

Step 3: Set Up Routing and Input Mapping

Ensure Telegram messages are classified and routed to the correct branch, then mapped into structured fields.

In Detect Link or Command, set the first condition to compare chat IDs using {{ $json.message.chat.id }} equals {{ $json.message.from.id }}.
Set the second condition in Detect Link or Command to check if {{ $json.message.text }} contains =https:// to detect article links.
In Assign Chat Input, create chatInput with value {{ $('Telegram Intake Trigger').item.json.message.text }}.
In Check Generate Command, keep the equality check between {{ $json.message.chat.id }} and {{ $json.message.from.id }}, and confirm the second condition checks if {{ $json.message.text }} contains generate.

Tip: This structure lets Detect Link or Command route links to summaries and route the “generate” command to the post generator.

Step 4: Set Up AI Insight Processing

Configure the AI agent to read the article and map its outputs into structured fields for storage and messaging.

In Insight Gatherer, set text to =Read the article {{ $json.chatInput }} and do as requiered by your role and keep promptType as define.
Attach OpenAI Chat Engine A as the language model for Insight Gatherer and ensure it uses model gpt-4.1-mini.
Credential Required: Connect your openAiApi credentials in OpenAI Chat Engine A. This model powers Insight Gatherer.
In Map Output Fields, map structured outputs using these expressions: Title {{$json.output.match(/Title:\s*([\s\S]*?)(?:\n|$)/)?.[1]}}, Link {{$json.output.extractUrl()}}, Summary {{$json.output.match(/Summary:\s*([^]+?)Insight & Commentary:/s)?.[1]?.trim()}}, and Insight & Commentary {{$json.output.match(/Insight & Commentary:\s*([^]+)/s)?.[1]}}.
In Normalize Records, keep the code as const fields = $input.all().map((item) => item.json); return fields; so the output matches the Google Sheets append format.

Step 5: Configure Output and Delivery

Send summaries and generated posts back to Telegram, and store summaries in Google Sheets.

Connect Map Output Fields → Normalize Records → Append Sheet Row to log insights to your sheet.
In Send Summary Message, set text to {{ $('Map Output Fields').item.json.Summary }} and chatId to {{ $('Telegram Intake Trigger').item.json.message.chat.id }}.
Credential Required: Connect your telegramApi credentials in Send Summary Message.
In Compose LinkedIn Post, keep the prompt text for post generation and ensure OpenAI Chat Engine B is attached as the language model.
Credential Required: Connect your openAiApi credentials in OpenAI Chat Engine B. This model powers Compose LinkedIn Post.
In Deliver LinkedIn Post, set text to {{ $json.output }} and chatId to {{ $('Telegram Intake Trigger').item.json.message.chat.id }}.
Credential Required: Connect your telegramApi credentials in Deliver LinkedIn Post.

⚠️ Common Pitfall: If Check Generate Command doesn’t match the word generate, Compose LinkedIn Post and Deliver LinkedIn Post will never run.

Step 6: Test and Activate Your Workflow

Validate the end-to-end flow from Telegram intake to summary storage and post delivery.

Click Execute Workflow and send a Telegram message containing a URL to trigger Telegram Intake Trigger and the summary path.
Confirm that Send Summary Message replies with a summary and that Append Sheet Row adds a new row in your Google Sheet.
Send a Telegram message containing the word generate to trigger Check Generate Command and confirm Deliver LinkedIn Post returns the generated post.
When successful, toggle the workflow to Active to run continuously in production.

🔒

Unlock Full Step-by-Step Guide

Get the complete implementation guide + downloadable template

Common Gotchas

Google Sheets credentials can expire or need specific permissions. If things break, check n8n’s Credentials panel and confirm the Google account still has edit access to the sheet.
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

How long does it take to set up this Telegram Sheets drafts automation?

About 30 minutes if your Telegram bot and Google Sheet are ready.

Do I need coding skills to automate Telegram Sheets drafts?

No. You’ll mostly connect accounts and copy/paste credentials. A little comfort editing text prompts helps, but it’s optional.

Is n8n free to use for this Telegram Sheets drafts workflow?

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 OpenAI (or your LLM) API costs, which are usually just a few dollars a month at low volume.

Where can I host n8n to run this automation?

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.

Can I customize this Telegram Sheets drafts workflow for a specific LinkedIn tone?

Yes, and you should. Update the prompt inside the “Compose LinkedIn Post” agent to match your voice (short punchy lines, more story-based, more technical, etc.). You can also change what gets stored by editing the “Map Output Fields” step, then adding or renaming columns in Google Sheets. Common tweaks include adding a “Topic” column, generating multiple hooks, or writing in a second language.

Why is my Telegram connection failing in this workflow?

Usually it’s the bot token. Recreate or re-copy the token from BotFather, then update the Telegram credentials used by the Telegram Trigger and send nodes in n8n. Also check that you’re messaging the correct bot, and that n8n can receive inbound webhooks (common issue on locked-down self-hosted setups). If it fails only sometimes, rate limits or long article fetch times can cause timeouts, so increasing timeouts in the HTTP request path can help.

How many articles can this Telegram Sheets drafts automation handle?

A lot. On n8n Cloud Starter, you’re typically fine for a personal workflow and a steady weekly pipeline, and bigger plans handle more volume. If you self-host, there’s no execution limit (it depends on your server). Practically, the bottleneck is AI and article fetching, so assume a handful of articles per minute, not hundreds.

Is this Telegram Sheets drafts automation better than using Zapier or Make?

Often, yes. This workflow relies on conditional routing (“link vs generate”), structured mapping, and AI agent steps, which is where n8n tends to feel more flexible and less paywalled. Zapier or Make can still do it, but multi-step AI workflows usually get expensive and harder to maintain. If you only want “save a link to Sheets,” those tools are fine. If you want the on-demand draft loop inside Telegram, n8n is a better fit. Talk to an automation expert if you’re stuck choosing.

Once this is running, your “I should post about this” moments stop evaporating. The workflow keeps the library organized and the drafts flowing, so you can focus on publishing and refining your point of view.

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.

Lisa Granqvist

Workflow Automation Expert

Expert in workflow automation and no-code tools.

{
"id": null,
"meta": {
"instanceId": "47d467ee5a1f255eeb1b14fbc56f101df6ed329b5349c9e23e0ed3eda45f62cf",
"templateId": null,
"templateCredsSetupCompleted": null
},
"name": "Automated LinkedIn Content Workflow",
"tags": [],
"nodes": [
{
"id": "flowpast-topbar-11279",
"name": "Flowpast Branding",
"type": "n8n-nodes-base.stickyNote",
"position": [
85,
-5
],
"parameters": {
"color": 7,
"width": 1580,
"height": 80,
"content": "## Flowpast.com | Automation Workflow Library\n**\ud83d\udcd6 Full tutorial & setup guide:** flowpast.com"
},
"typeVersion": 1
},
{
"id": "db3a4a3d-61b5-4563-85a3-f1d6f873caf7",
"name": "Map Output Fields",
"type": "n8n-nodes-base.set",
"position": [
965,
205
],
"parameters": {
"options": [],
"assignments": {
"assignments": [
{
"id": "6232fba4-479d-4d09-9500-55fbaf47cb2c",
"name": "=Title",
"type": "string",
"value": "={{$json.output.match(/Title:\\s*([\\s\\S]*?)(?:\\n|$)/)?.[1]}}"
},
{
"id": "5aaf55f5-cecc-4545-83c7-0250d6322005",
"name": "Link",
"type": "string",
"value": "={{$json.output.extractUrl()}}"
},
{
"id": "f4ff9150-3496-4ead-b5af-d23afbfeb7b5",
"name": "Summary",
"type": "string",
"value": "={{$json.output.match(/Summary:\\s*([^]+?)Insight & Commentary:/s)?.[1]?.trim()}}"
},
{
"id": "b866a8cb-a4da-45cc-87af-8f9801c57970",
"name": "Insight & Commentary",
"type": "string",
"value": "={{$json.output.match(/Insight & Commentary:\\s*([^]+)/s)?.[1]}}"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "41ab2eb9-3208-454b-b90e-69cdad1696bc",
"name": "Insight Gatherer",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
745,
245
],
"parameters": {
"text": "=Read the article {{ $json.chatInput }} and do as requiered by your role",
"options": {
"systemMessage": "=Today is: {{ $now.format('yyyy-MM-dd') }}\nArticle link: {{ $json.chatInput }}\n\nROLE: \nYou are a sharp research and insight assistant. Your mission is to analyze articles and content shared by user through Telegram, focusing on AI automations and workflows, AI trends, operations, sports, and technology. Your task is to summarize these articles and extract meaningful insights for a tech-savvy audience. Store your summaries and recommendations in an Airtable.\n\nInfuse user\u2019s voice: a blend of formal, direct, and witty. Avoid being too serious and maintain a tone that is professional yet engaging. Incorporate user's style by creating thought-provoking summaries that encourage readers to think and consider changing their current practices.\n\nCAPABILITIES:\n\n- Receive articles from user via Telegram.\n- Access, read, and interpret the full content of each article.\n- Identify key points, themes, and innovations relevant to AI, operations, and technology.\n- Provide concise, insightful summaries in a formal, direct, and witty tone.\n- Add user\u2019s perspective and commentary, using a style that is short, to the point, and avoids unnecessary seriousness.\n- Log the summaries and insights into Airtable with fields: Title, Link, Summary, Insight & Commentary.\n\nINPUT HANDLING:\n\n- Upon receiving a link:\n  - Visit the page and read the full article.\n  - Extract the article\u2019s title and URL.\n  - Write a concise summary, focusing on AI automation, efficiency, and innovation.\n  - Under \"Insight & Commentary,\" provide a witty analysis reflective of user\u2019s style and perspective.\n\n- If given text without a link:\n  - Use the entire text as the Title.\n  - Write a summary of the main ideas.\n  - Provide commentary that ties back to user\u2019s core themes and perspectives.\n\nRESPONSE GUIDELINES:\n\n- Be concise, direct, and insightful.\n- Maintain a thought-leadership tone: confident and witty.\n- Focus on solutions, innovation, and efficiency.\n- Avoid politics and overly serious topics.\n\nLIMITATIONS:\n\n- If the article is inaccessible, note this clearly in the Summary field.\n- If the article is already logged (duplicate), reply with \u201cYou already shared this with me, please send a different one.\u201d\n- Skip irrelevant content and avoid repeating the same commentary across entries.\n- Do not delve into politics or overly serious personal subjects.\n\nEXAMPLE OUTPUT:\n\nTitle: \"AI Trends Transforming Operations in 2023\"\n\nLink: https://example.com/ai-trends-2023\n\nSummary: The article explores how AI trends, including automated workflows and intelligent systems, are revolutionizing operational efficiency across industries. Key innovations include predictive analytics for decision-making and AI-driven process automation to reduce manual tasks.\n\nInsight & Commentary: \"Imagine your workflow humming along like a well-oiled machine, where AI takes the mundane off your plate. That's not just the future\u2014it's happening now. Embrace these trends or risk being left behind. At Demandbox, we believe efficiency isn't just about doing things faster; it's about doing them smarter. Let's lead this change.\"\n"
},
"promptType": "define"
},
"typeVersion": 2.1
},
{
"id": "a7623437-eaea-4532-8095-c9e8cbb7b549",
"name": "Telegram Intake Trigger",
"type": "n8n-nodes-base.telegramTrigger",
"position": [
85,
215
],
"webhookId": "5f4df1e8-8c19-43bb-a4af-1a1634026f26",
"parameters": {
"updates": [
"message"
],
"additionalFields": []
},
"credentials": {
"telegramApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1.2
},
{
"id": "3ca47971-d78c-426d-8606-67bd8c9d43f8",
"name": "Send Summary Message",
"type": "n8n-nodes-base.telegram",
"position": [
1565,
265
],
"webhookId": "d5198b06-26a5-4b34-927a-d41db17b198b",
"parameters": {
"text": "={{ $('Map Output Fields').item.json.Summary }}",
"chatId": "={{ $('Telegram Intake Trigger').item.json.message.chat.id }}",
"additionalFields": {
"appendAttribution": false
}
},
"credentials": {
"telegramApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1.2
},
{
"id": "8517de48-6fd4-4eec-aa2e-ef860bf05680",
"name": "Check Generate Command",
"type": "n8n-nodes-base.if",
"position": [
535,
445
],
"parameters": {
"options": [],
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "fb91ed8a-73f1-4f9f-bae1-5ebbae9a4501",
"operator": {
"type": "number",
"operation": "equals"
},
"leftValue": "={{ $json.message.chat.id }}",
"rightValue": "={{ $json.message.from.id }}"
},
{
"id": "f30da9f0-a56c-4bae-a73f-0517b521c11a",
"operator": {
"type": "string",
"operation": "contains"
},
"leftValue": "={{ $json.message.text }}",
"rightValue": "generate"
}
]
}
},
"typeVersion": 2.2
},
{
"id": "600d6330-cd34-4c03-9486-12022aeda73c",
"name": "Normalize Records",
"type": "n8n-nodes-base.code",
"position": [
1165,
255
],
"parameters": {
"jsCode": "const fields = $input.all().map((item) => item.json);\nreturn fields;\n"
},
"typeVersion": 2
},
{
"id": "422e62e0-06ac-482c-b345-956fe22e712f",
"name": "Detect Link or Command",
"type": "n8n-nodes-base.if",
"position": [
305,
255
],
"parameters": {
"options": [],
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "dc4561f4-6560-4e5d-b540-e5e0b7aef3e7",
"operator": {
"type": "number",
"operation": "equals"
},
"leftValue": "={{ $json.message.chat.id }}",
"rightValue": "={{ $json.message.from.id }}"
},
{
"id": "469202b0-eac6-48c9-a74f-76607dd8d6e2",
"operator": {
"type": "string",
"operation": "contains"
},
"leftValue": "={{ $json.message.text }}",
"rightValue": "=https://"
}
]
}
},
"typeVersion": 2.2
},
{
"id": "99cd2918-f4df-477e-ad9b-a0bcf245346d",
"name": "Assign Chat Input",
"type": "n8n-nodes-base.set",
"position": [
515,
205
],
"parameters": {
"options": [],
"assignments": {
"assignments": [
{
"id": "9afa2e7f-0c53-462d-ae0c-66b622787297",
"name": "=chatInput",
"type": "string",
"value": "={{ $('Telegram Intake Trigger').item.json.message.text }}"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "b26e0137-cb6a-44c1-bf30-64b8ef5705fc",
"name": "Append Sheet Row",
"type": "n8n-nodes-base.googleSheets",
"position": [
1365,
205
],
"parameters": {
"operation": "append",
"sheetName": {
"__rl": true,
"mode": "list",
"value": ""
},
"documentId": {
"__rl": true,
"mode": "list",
"value": ""
}
},
"credentials": {
"googleSheetsOAuth2Api": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 4.7
},
{
"id": "3e6333d6-ea90-4874-8f07-c2a0bf9b9c6a",
"name": "Retrieve Sheet Rows",
"type": "n8n-nodes-base.googleSheetsTool",
"position": [
1085,
665
],
"parameters": {
"sheetName": {
"__rl": true,
"mode": "list",
"value": ""
},
"documentId": {
"__rl": true,
"mode": "list",
"value": ""
}
},
"credentials": {
"googleSheetsOAuth2Api": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 4.7
},
{
"id": "58edfc02-05a3-4bce-a91f-5b54125e470b",
"name": "Deliver LinkedIn Post",
"type": "n8n-nodes-base.telegram",
"position": [
1045,
445
],
"webhookId": "8ec00888-6a05-48a2-8087-0b78261ad4fd",
"parameters": {
"text": "={{ $json.output }}",
"chatId": "={{ $('Telegram Intake Trigger').item.json.message.chat.id }}",
"additionalFields": {
"appendAttribution": false
}
},
"credentials": {
"telegramApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1.2
},
{
"id": "209e9e99-fb2c-4d8b-951d-faaace17ff92",
"name": "Compose LinkedIn Post",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
795,
485
],
"parameters": {
"text": "Using the information you have, create a well written viral LinkedIn post ",
"options": {
"systemMessage": "=ROLE: \nYou are a LinkedIn content assistan. Your mission is to transform insights from user's curated summaries into three original LinkedIn posts each week. Craft posts that are educational, engaging, and reflect user\u2019s voice and vision on AI automations, workflows, and technological innovation.\n\nuser\u2019s style is formal, direct, and witty\u2014short and to the point. Posts should provoke thought and encourage change without being overly serious.\n\nCAPABILITIES:\n- When asked to generate a LinkedIn post, review the recent articles in the AirTable and generate the post. \n- Scan the Airtable for unused insights and summaries.\n- Evaluate entries to determine their potential to resonate with a tech-savvy audience.\n- Generate three different LinkedIn posts weekly, grounded in user\u2019s style and voice.\n- Ensure posts align with themes of AI trends, operations, and technology while being thought-leadership focused.\n- Flag used content to prevent duplication.\n\nINPUT HANDLING:\n\n- Identify new content: Query Airtable for unused summaries. If none are available, find recent articles related to user\u2019s themes.\n- Select compelling topics based on novelty and relevance.\n- Compose three posts per week:\n  - Start with an engaging hook.\n  - Develop the core argument in concise paragraphs.\n  - Incorporate user\u2019s beliefs and encourage readers to think differently.\n  - End with a thought-provoking statement or call to action.\n\nRESPONSE GUIDELINES:\n\n- Voice: Write in the first person as user (\"I\"/\"we\"). Maintain a confident, direct tone with a touch of wit.\n- Length: Aim for ~150\u2013250 words per post. Use short sentences and paragraphs for clarity.\n- Content: Focus on AI, technology, and operations. Avoid politics and overly serious topics.\n- Accuracy: Never invent facts. Use reputable sources for supporting data. If data is uncertain, omit it.\n\nLIMITATIONS:\n- Before you generate the LinkedIn post, you MUST check first \"the rules tool\" and take into consideration everything is written inside, those are your do and don't guideliness for you to keep in mind. \n- Do not reuse content already posted on LinkedIn.\n- Avoid personal anecdotes and politics.\n- If Airtable has no new entries, inform user and suggest searching for fresh articles.\n\nEXAMPLE OUTPUT:\n\nPost 1: \n\"AI automations are not just a trend\u2014they're a necessity. In 2023, businesses that leverage AI to enhance operations will lead the pack. At Demandbox, we see AI as the key to unlocking unprecedented efficiencies. It's time to rethink how we work\u2014let's innovate, optimize, and transform.\"\n\nPost 2: \n\"Imagine a world where operations run smoothly, effortlessly\u2014powered by AI. That's the future we're building at Demandbox. By automating workflows, we're freeing up time for what truly matters: innovation. How are you using AI to reshape your operations?\"\n\nPost 3: \n\"Efficiency and innovation go hand in hand. As we embrace AI trends, let's focus on improving the way we work. At Demandbox, we're committed to making tech more than just a tool\u2014it's a game-changer. Join us as we redefine efficiency in the tech world.\"\n"
},
"promptType": "define"
},
"typeVersion": 2.1
},
{
"id": "ab17ec27-9393-49cd-bdad-6df3390f5506",
"name": "OpenAI Chat Engine A",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
745,
95
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "gpt-4.1-mini"
},
"options": [],
"builtInTools": []
},
"credentials": {
"openAiApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1.3
},
{
"id": "a265b2fb-ca43-4b4e-9b6e-9a34bb249111",
"name": "OpenAI Chat Engine B",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
825,
705
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "gpt-4.1-mini"
},
"options": [],
"builtInTools": []
},
"credentials": {
"openAiApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1.3
}
],
"active": false,
"pinData": [],
"settings": {
"executionOrder": "v1"
},
"versionId": null,
"connections": {
"Assign Chat Input": {
"main": [
[
{
"node": "Insight Gatherer",
"type": "main",
"index": 0
}
]
]
},
"Normalize Records": {
"main": [
[
{
"node": "Append Sheet Row",
"type": "main",
"index": 0
}
]
]
},
"Map Output Fields": {
"main": [
[
{
"node": "Normalize Records",
"type": "main",
"index": 0
}
]
]
},
"Check Generate Command": {
"main": [
[
{
"node": "Compose LinkedIn Post",
"type": "main",
"index": 0
}
],
[]
]
},
"Telegram Intake Trigger": {
"main": [
[
{
"node": "Detect Link or Command",
"type": "main",
"index": 0
}
]
]
},
"Insight Gatherer": {
"main": [
[
{
"node": "Map Output Fields",
"type": "main",
"index": 0
}
]
]
},
"OpenAI Chat Engine A": {
"ai_languageModel": [
[
{
"node": "Insight Gatherer",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"OpenAI Chat Engine B": {
"ai_languageModel": [
[
{
"node": "Compose LinkedIn Post",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Append Sheet Row": {
"main": [
[
{
"node": "Send Summary Message",
"type": "main",
"index": 0
}
]
]
},
"Detect Link or Command": {
"main": [
[
{
"node": "Assign Chat Input",
"type": "main",
"index": 0
}
],
[
{
"node": "Check Generate Command",
"type": "main",
"index": 0
}
]
]
},
"Compose LinkedIn Post": {
"main": [
[
{
"node": "Deliver LinkedIn Post",
"type": "main",
"index": 0
}
]
]
},
"Retrieve Sheet Rows": {
"ai_tool": [
[
{
"node": "Compose LinkedIn Post",
"type": "ai_tool",
"index": 0
}
]
]
}
}
}