January 22, 2026

OpenAI to Telegram, your daily AI news digest

Lisa Granqvist Partner Workflow Automation Expert

Get a free AI assessment → ⬇️ Use template

Your morning “AI catch-up” starts with good intentions, then turns into 12 open tabs, half-read headlines, and that nagging feeling you missed the one story that actually matters. By the time you’ve skimmed it all, your day is already gone.

This AI news digest automation hits marketing leads and founders hardest because you need signal, not noise. But analysts and agency teams feel it too. The outcome is simple: a clean, ranked Telegram brief every morning, ready to forward to your team.

Below, you’ll see how the workflow pulls from multiple sources, uses OpenAI to summarize and format the top stories, generates cover art, and posts the digest to Telegram (plus a chat mode for follow-up questions).

How This Automation Works

See how this solves the problem:

n8n Workflow Template: OpenAI to Telegram, your daily AI news digest

Click to explore

flowchart LR

    subgraph sg0["Daily 8 AM Flow"]
        direction LR
        n0@{ icon: "mdi:play-circle", form: "rounded", label: "Daily 8 AM Trigger", pos: "b", h: 48 }
        n1@{ icon: "mdi:swap-vertical", form: "rounded", label: "Workflow Configuration", 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/httprequest.dark.svg' width='40' height='40' /></div><br/>Fetch AI News Source 1"]
        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/httprequest.dark.svg' width='40' height='40' /></div><br/>Fetch AI News Source 2"]
        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/>Fetch AI News Source 3"]
        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/>Aggregate and Rank News"]
        n6@{ icon: "mdi:robot", form: "rounded", label: "AI News Summarizer Agent", pos: "b", h: 48 }
        n7@{ icon: "mdi:brain", form: "rounded", label: "OpenAI Chat Model", pos: "b", h: 48 }
        n8@{ icon: "mdi:robot", form: "rounded", label: "Structured Output Parser", pos: "b", h: 48 }
        n9@{ icon: "mdi:robot", form: "rounded", label: "Generate AI Image", 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/code.svg' width='40' height='40' /></div><br/>Format Telegram Message"]
        n11["<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 to Telegram"]
        n9 --> n10
        n7 -.-> n6
        n0 --> n1
        n2 --> n5
        n3 --> n5
        n4 --> n5
        n1 --> n2
        n1 --> n3
        n1 --> n4
        n5 --> n6
        n10 --> n11
        n6 --> n9
        n8 -.-> n6
    end

    subgraph sg1["Telegram Flow"]
        direction LR
        n12@{ icon: "mdi:robot", form: "rounded", label: "Chat AI Agent", pos: "b", h: 48 }
        n13@{ icon: "mdi:brain", form: "rounded", label: "OpenAI Chat Model for Chat", pos: "b", h: 48 }
        n14@{ icon: "mdi:memory", form: "rounded", label: "Window Buffer Memory", pos: "b", h: 48 }
        n15["<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"]
        n15 --> n12
        n14 -.-> n12
        n13 -.-> n12
    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 n0,n15 trigger
    class n6,n8,n9,n12 ai
    class n7,n13 aiModel
    class n14 ai
    class n2,n3,n4 api
    class n5,n10 code
    classDef customIcon fill:none,stroke:none
    class n2,n3,n4,n5,n10,n11,n15 customIcon

The Challenge: Keeping Up With AI News Without Losing Your Morning

AI news moves fast, and it’s scattered on purpose. One story breaks on a tech blog, the reaction happens on social, and the useful analysis is buried in a newsletter you’ll read “later.” Manually checking sources feels manageable until it becomes a daily habit that steals focus. You also end up forwarding messy links to your team with zero context, which sparks more questions than clarity. Honestly, it’s not the reading that kills you. It’s the switching, sorting, and summarizing when you should be planning, writing, or selling.

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

You scan the same story three times because each site rewrites the headline slightly.
Important updates get missed because nobody “owns” the daily roundup.
Sharing to Telegram is inconsistent, so the team stops trusting the channel.
By the time you summarize the news, it’s already lunchtime.

The Fix: An OpenAI-Powered Daily Digest Posted to Telegram

This workflow turns “random browsing” into a repeatable briefing. Every morning at a scheduled time, n8n fetches fresh items from multiple RSS feeds (Google News AI, The Verge, and TechCrunch in the default setup). It then combines everything, removes duplicates, and ranks stories by recency so you only deal with the top five. Next, an AI Agent powered by OpenAI rewrites the headlines and summaries into a digestible format that reads like a brief, not a dump of links. A separate OpenAI step generates a futuristic cover image using DALL·E 3, and the workflow composes a clean Telegram post and publishes it to your chosen chat.

The flow starts with the morning schedule trigger and ends with a formatted Telegram message your team can scan in under two minutes. There’s also a second branch: a Telegram incoming trigger that routes messages into a “conversation” AI Agent, so you can ask follow-up questions about what you just read.

What Changes: Before vs. After

What This Eliminates

Impact You’ll See

Manually checking three or more news sources every morning.
Copying links into Telegram and writing rushed one-line summaries.
Reading duplicates because different outlets syndicate the same story.
Starting from scratch each day with no consistent digest format.

Most teams get about 1 hour back every workday.
A consistent “single brief” that’s easy to skim and forward.
Cleaner internal alignment because everyone sees the same top five stories.
Fewer “did you see this?” pings scattered across chats.
A built-in way to ask follow-up questions without opening more tabs.

Real-World Impact

Say you check 3 sources each morning and spend maybe 10 minutes per source between skimming, opening links, and deciding what’s worth sharing. That’s about 30 minutes, plus another 10 minutes to format a message your team will actually read. With this workflow, you spend about 2 minutes glancing at the Telegram digest, then ask one follow-up question in chat mode if needed. You’re buying back roughly 30 to 40 minutes a day, without losing coverage.

Requirements

n8n instance (try n8n Cloud free)
Self-hosting option if you prefer (Hostinger works well)
OpenAI account for summaries and DALL·E images.
Telegram to deliver the daily digest to a chat.
OpenAI API key (get it from the OpenAI dashboard)

Skill level: Intermediate. You’ll connect credentials, paste a Telegram chat ID, and adjust a couple of workflow variables.

Need help implementing this? Talk to an automation expert (free 15-minute consultation).

The Workflow Flow

A scheduled morning trigger kicks everything off. At the set time (8:00 AM in the default workflow), n8n starts the run and loads your saved variables like RSS URLs and your Telegram chat ID.

News gets pulled from multiple sources. Three HTTP requests fetch RSS feeds (Google News AI, The Verge, TechCrunch). You can swap these URLs to track other topics, or add more feeds if you want broader coverage.

The workflow cleans and ranks the stories. A code step combines items, removes duplicates, and selects the top five based on recency so you don’t end up with repetitive links.

OpenAI creates the “brief,” then Telegram delivers it. An AI Agent writes concise summaries, DALL·E generates header art, and another code step formats everything into a Telegram-ready message before posting it to your chosen chat.

You can easily modify the news sources and summary tone to fit your audience. See the full implementation guide below for customization options.

Step-by-Step Implementation Guide

Step 1: Configure the Scheduled Trigger

Set the daily schedule that kicks off the morning news digest.

Open Scheduled Morning Trigger and set the schedule rule to run at 8 using the interval option.
Confirm it connects to Setup Variables as the next node in the main flow.

Step 2: Define Source URLs and Telegram Target

Store the RSS sources and Telegram chat target used across the workflow.

In Setup Variables, set newsSource1Url to https://news.google.com/rss/search?q=artificial+intelligence&hl=en-US&gl=US&ceid=US:en.
Set newsSource2Url to https://www.theverge.com/rss/ai-artificial-intelligence/index.xml.
Set newsSource3Url to https://techcrunch.com/category/artificial-intelligence/feed/.
Set telegramChatId to your Telegram chat ID (replace <__PLACEHOLDER_VALUE__Your Telegram Chat ID__> with your real ID).

⚠️ Common Pitfall: If the Telegram chat ID is wrong, Dispatch Telegram Post will fail even though the workflow runs successfully.

Step 3: Fetch News in Parallel

The workflow pulls three RSS feeds simultaneously for speed and resilience.

In Retrieve News Feed A, set URL to {{ $('Setup Variables').first().json.newsSource1Url }}.
In Retrieve News Feed B, set URL to {{ $('Setup Variables').first().json.newsSource2Url }}.
In Retrieve News Feed C, set URL to {{ $('Setup Variables').first().json.newsSource3Url }}.
Ensure Setup Variables outputs to Retrieve News Feed A, Retrieve News Feed B, and Retrieve News Feed C in parallel.

Setup Variables outputs to both Retrieve News Feed A and Retrieve News Feed B and Retrieve News Feed C in parallel.

Step 4: Aggregate, Rank, and Summarize Stories with AI

Combine all articles, rank them by recency, and generate structured summaries.

In Combine & Rank Stories, keep the existing JavaScript to parse RSS/JSON, deduplicate, and select the top 5 stories.
In AI Digest Summarizer, set Text to You are an AI news summarizer. Analyze the top 5 AI news stories provided and create concise, engaging summaries for each. Return the data in the structured format defined by the output parser. Include story titles, summaries, and links..
Open Structured Result Parser and keep the JSON Schema Example that includes stories and imagePrompt.
Connect OpenAI Chat Engine as the language model for AI Digest Summarizer and ensure the model is gpt-4.1-mini.

Credential Required: Connect your openAiApi credentials to OpenAI Chat Engine.

Credential Required: Connect your openAiApi credentials to Create AI Cover Art.

Tip:

For AI tool sub-nodes like Structured Result Parser, add credentials to the parent node (OpenAI Chat Engine connected to AI Digest Summarizer), not the parser itself.

Step 5: Generate the Cover Art and Format the Telegram Digest

Create the AI image and format the digest message before sending to Telegram.

In Create AI Cover Art, set Prompt to {{ $json.output.imagePrompt }} and keep Resource as image.
In Compose Telegram Digest, keep the JavaScript that builds the message and maps telegramMessage and imageUrl for the next node.
Verify the flow order: AI Digest Summarizer → Create AI Cover Art → Compose Telegram Digest.

Step 6: Configure Telegram Output and Interactive Chat

Send the digest to Telegram and enable an inbound AI conversation trigger.

In Dispatch Telegram Post, set Operation to sendPhoto.
Set File to {{ $json.imageUrl }} and Chat ID to {{ $('Setup Variables').first().json.telegramChatId }}.
Set Caption to {{ $json.telegramMessage }} and Parse Mode to Markdown.
In Telegram Incoming Trigger, leave Updates set to message and connect it to Conversation AI Agent.
Connect Chat Model Engine to Conversation AI Agent and keep the model set to gpt-4.1-mini.
Attach Chat Window Memory to Conversation AI Agent with Context Window Length set to 40.

Credential Required: Connect your telegramApi credentials to Dispatch Telegram Post.

Credential Required: Connect your telegramApi credentials to Telegram Incoming Trigger.

Credential Required: Connect your openAiApi credentials to Chat Model Engine.

Tip:

For AI tool sub-nodes like Chat Window Memory, add credentials to the parent node (Chat Model Engine connected to Conversation AI Agent), not the memory node.

Step 7: Test and Activate Your Workflow

Validate both the scheduled digest and the interactive chat response before enabling production use.

Click Execute Workflow to run from Scheduled Morning Trigger and verify that Dispatch Telegram Post sends a formatted photo message.
Trigger Telegram Incoming Trigger by sending a message to your bot and confirm Conversation AI Agent replies.
Successful execution should show a Telegram message with a cover image, five story summaries, and clickable links.
When satisfied, toggle the workflow Active to enable the daily schedule and live chat handling.

🔒

Unlock Full Step-by-Step Guide

Get the complete implementation guide + downloadable template

Watch Out For

Telegram credentials can expire or the bot can lose access to the chat. If things break, check the bot is still in the chat and confirm the Chat ID in your workflow variables first.
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.

Common Questions

How quickly can I implement this AI news digest automation?

About 30 minutes if you already have your OpenAI key and a Telegram bot.

Can non-technical teams implement this AI news digest?

Yes. You won’t write code, but you will connect accounts and paste in a Telegram chat ID.

Is n8n free to use for this AI news digest 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 API costs for the summarizer and DALL·E image generation.

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.

How do I adapt this AI news digest automation solution to my specific challenges?

You can swap the RSS sources by editing the feed URLs stored in the workflow’s setup/variable step (the node labeled “Setup Variables”). For the writing style, update the system instructions inside the “AI Digest Summarizer” agent so it matches your voice (more technical, more casual, “TL;DR only,” and so on). If you want more than five stories, adjust the ranking step that selects the top items before the AI summary runs. And if Telegram isn’t the final destination, you can replace the “Dispatch Telegram Post” step with email, Slack, or a database write.

Why is my Telegram connection failing in this workflow?

Most of the time it’s the basics: the bot token is wrong, the bot was removed from the chat, or the chat ID is pointing to a different conversation. Regenerate the Telegram credential if needed, then confirm the saved telegramChatId value matches the numeric ID for the target chat. If you’re posting to a group, make sure the bot has permission to send messages there.

What’s the capacity of this AI news digest solution?

For a daily digest, capacity is effectively a non-issue because it runs once per day and processes a small batch of items (top five by default). If you scale it to hourly updates or add many feeds, n8n Cloud plan limits and OpenAI rate limits start to matter, but self-hosting removes execution caps and shifts the bottleneck to your server.

Is this AI news digest automation better than using Zapier or Make?

Often, yes, because this workflow benefits from richer logic (deduping, ranking, structured parsing, and an interactive chat branch) without turning into an expensive tangle of steps. n8n also gives you the option to self-host, which is a big deal if you want predictable costs. Zapier or Make can still be fine if you only need a simple RSS-to-Telegram relay with a basic summary. The moment you care about quality and consistency, you’ll want the extra control. Talk to an automation expert if you want help choosing.

Once this is running, your “stay informed” habit stops stealing attention. The workflow handles the repeatable briefing so you can use your mornings for work that actually moves things forward.

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": "57ce30275d453cb60e7a9318323f8f800eb7a8116ffc80ed687d83948d2e0252",
"templateId": null,
"templateCredsSetupCompleted": null
},
"name": "Automated AI News Digest Delivery",
"tags": [],
"nodes": [
{
"id": "flowpast-topbar-12048",
"name": "Flowpast Branding",
"type": "n8n-nodes-base.stickyNote",
"position": [
0,
20
],
"parameters": {
"color": 7,
"width": 880,
"height": 80,
"content": "## Flowpast.com | Automation Workflow Library\n**\ud83d\udcd6 Full tutorial & setup guide:** flowpast.com"
},
"typeVersion": 1
},
{
"id": "80ab307e-e246-49ed-92f7-eee1c496f1b0",
"name": "Scheduled Morning Trigger",
"type": "n8n-nodes-base.scheduleTrigger",
"position": [
770,
270
],
"parameters": {
"rule": {
"interval": [
{
"triggerAtHour": 8
}
]
}
},
"typeVersion": 1.2
},
{
"id": "b4f855a8-69b1-42a6-9bf3-27688b21fcde",
"name": "Setup Variables",
"type": "n8n-nodes-base.set",
"position": [
560,
300
],
"parameters": {
"options": [],
"assignments": {
"assignments": [
{
"id": "id-1",
"name": "newsSource1Url",
"type": "string",
"value": "https://news.google.com/rss/search?q=artificial+intelligence&hl=en-US&gl=US&ceid=US:en"
},
{
"id": "id-2",
"name": "newsSource2Url",
"type": "string",
"value": "https://www.theverge.com/rss/ai-artificial-intelligence/index.xml"
},
{
"id": "id-3",
"name": "newsSource3Url",
"type": "string",
"value": "https://techcrunch.com/category/artificial-intelligence/feed/"
},
{
"id": "id-4",
"name": "telegramChatId",
"type": "string",
"value": "<__PLACEHOLDER_VALUE__Your Telegram Chat ID__>"
}
]
},
"includeOtherFields": true
},
"typeVersion": 3.4
},
{
"id": "271dcf62-755d-4ebc-96cf-7c084d140fc2",
"name": "Retrieve News Feed A",
"type": "n8n-nodes-base.httpRequest",
"position": [
370,
150
],
"parameters": {
"url": "={{ $('Setup Variables').first().json.newsSource1Url }}",
"options": {
"response": {
"response": {
"neverError": true
}
}
}
},
"typeVersion": 4.2,
"continueOnFail": true
},
{
"id": "e030ed22-de8f-4904-ad13-f9e75c22ef9a",
"name": "Retrieve News Feed B",
"type": "n8n-nodes-base.httpRequest",
"position": [
350,
300
],
"parameters": {
"url": "={{ $('Setup Variables').first().json.newsSource2Url }}",
"options": {
"response": {
"response": {
"neverError": true
}
}
}
},
"typeVersion": 4.2,
"continueOnFail": true
},
{
"id": "49ff1fe2-71e9-41bf-94e9-482c06c25a8c",
"name": "Retrieve News Feed C",
"type": "n8n-nodes-base.httpRequest",
"position": [
380,
450
],
"parameters": {
"url": "={{ $('Setup Variables').first().json.newsSource3Url }}",
"options": {
"response": {
"response": {
"neverError": true
}
}
}
},
"typeVersion": 4.2,
"continueOnFail": true
},
{
"id": "9554da09-b7b4-416a-8367-88297e6f7960",
"name": "Combine & Rank Stories",
"type": "n8n-nodes-base.code",
"position": [
170,
300
],
"parameters": {
"jsCode": "// Aggregate all news items from the three sources\nconst allNews = [];\n\n// Helper function to extract text from XML-like content\nfunction extractText(content, tag) {\n  const regex = new RegExp(`<${tag}[^>]*>([\\\\s\\\\S]*?)</${tag}>`, 'i');\n  const match = content.match(regex);\n  if (match) {\n    // Remove CDATA wrapper if present\n    let text = match[1].replace(/<!\\[CDATA\\[([\\s\\S]*?)\\]\\]>/g, '$1');\n    // Remove HTML tags\n    text = text.replace(/<[^>]+>/g, '');\n    return text.trim();\n  }\n  return '';\n}\n\n// Helper function to extract link from RSS item\nfunction extractLink(itemContent) {\n  // Try standard link tag\n  let link = extractText(itemContent, 'link');\n  if (link) return link;\n  \n  // Try guid if it looks like a URL\n  let guid = extractText(itemContent, 'guid');\n  if (guid && guid.startsWith('http')) return guid;\n  \n  return '';\n}\n\n// Collect items from all input sources\nfor (const item of $input.all()) {\n  const data = item.json;\n  \n  // Skip error responses\n  if (!data || data.error) continue;\n  \n  // Check if it's RSS/XML format (string response)\n  if (typeof data === 'string' || data.data) {\n    const xmlContent = typeof data === 'string' ? data : data.data;\n    if (typeof xmlContent === 'string' && xmlContent.includes('<item>')) {\n      // Parse RSS items\n      const itemRegex = /<item>([\\s\\S]*?)<\\/item>/gi;\n      let match;\n      while ((match = itemRegex.exec(xmlContent)) !== null) {\n        const itemContent = match[1];\n        const title = extractText(itemContent, 'title');\n        const description = extractText(itemContent, 'description') || extractText(itemContent, 'content:encoded');\n        const link = extractLink(itemContent);\n        const pubDate = extractText(itemContent, 'pubDate');\n        const source = extractText(itemContent, 'source') || 'RSS Feed';\n        \n        if (title) {\n          allNews.push({\n            title: title,\n            description: description.substring(0, 300) + (description.length > 300 ? '...' : ''),\n            link: link,\n            source: source,\n            publishedAt: pubDate || new Date().toISOString()\n          });\n        }\n      }\n    }\n  }\n  // Check if it's JSON API format (NewsAPI style)\n  else if (data.articles && Array.isArray(data.articles)) {\n    data.articles.forEach(article => {\n      allNews.push({\n        title: article.title || '',\n        description: (article.description || article.summary || '').substring(0, 300),\n        link: article.url || article.link || '',\n        source: article.source?.name || article.source || 'Unknown',\n        publishedAt: article.publishedAt || article.pubDate || new Date().toISOString()\n      });\n    });\n  }\n  // Check if it's a single article object\n  else if (data.title) {\n    allNews.push({\n      title: data.title || '',\n      description: (data.description || data.summary || '').substring(0, 300),\n      link: data.url || data.link || '',\n      source: data.source?.name || data.source || 'Unknown',\n      publishedAt: data.publishedAt || data.pubDate || new Date().toISOString()\n    });\n  }\n}\n\n// Remove duplicates based on title\nconst seen = new Set();\nconst uniqueNews = allNews.filter(item => {\n  const key = item.title.toLowerCase().trim();\n  if (seen.has(key)) return false;\n  seen.add(key);\n  return true;\n});\n\n// Rank by recency (most recent first)\nuniqueNews.sort((a, b) => {\n  const dateA = new Date(a.publishedAt);\n  const dateB = new Date(b.publishedAt);\n  return dateB - dateA;\n});\n\n// Get top 5 stories\nconst top5News = uniqueNews.slice(0, 5);\n\n// Return the top 5 stories\nreturn top5News.map(article => ({\n  json: {\n    title: article.title,\n    description: article.description,\n    link: article.link,\n    source: article.source,\n    publishedAt: article.publishedAt\n  }\n}));"
},
"typeVersion": 2
},
{
"id": "d3903770-607d-4be5-b299-c79c97bf7e5a",
"name": "AI Digest Summarizer",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
0,
290
],
"parameters": {
"text": "You are an AI news summarizer. Analyze the top 5 AI news stories provided and create concise, engaging summaries for each. Return the data in the structured format defined by the output parser. Include story titles, summaries, and links.",
"options": [],
"promptType": "define",
"hasOutputParser": true
},
"typeVersion": 2.2
},
{
"id": "58fba4c9-4c2f-4e8d-913d-c6056166cd6f",
"name": "OpenAI Chat Engine",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
50,
120
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "gpt-4.1-mini"
},
"options": []
},
"credentials": {
"openAiApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1.2
},
{
"id": "b0a0ac4c-d2d6-46b8-bc65-8061c717ebf0",
"name": "Structured Result Parser",
"type": "@n8n/n8n-nodes-langchain.outputParserStructured",
"position": [
200,
120
],
"parameters": {
"jsonSchemaExample": "{\n\t\"stories\": [\n\t\t{\n\t\t\t\"title\": \"AI Breakthrough in Healthcare\",\n\t\t\t\"summary\": \"Researchers develop new AI model for disease detection\",\n\t\t\t\"link\": \"https://example.com/article1\"\n\t\t},\n\t\t{\n\t\t\t\"title\": \"OpenAI Releases New Model\",\n\t\t\t\"summary\": \"Latest GPT model shows improved reasoning\",\n\t\t\t\"link\": \"https://example.com/article2\"\n\t\t}\n\t],\n\t\"imagePrompt\": \"A futuristic AI-themed illustration representing today's top AI news\"\n}"
},
"typeVersion": 1.3
},
{
"id": "e7753d4f-9eef-40d9-a585-8ee716cd63f2",
"name": "Create AI Cover Art",
"type": "@n8n/n8n-nodes-langchain.openAi",
"position": [
200,
490
],
"parameters": {
"prompt": "={{ $json.output.imagePrompt }}",
"options": {
"size": "1024x1024",
"dalleQuality": "standard"
},
"resource": "image"
},
"credentials": {
"openAiApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1.8
},
{
"id": "29f0025e-0a06-418a-89ef-8ad6caaa3567",
"name": "Compose Telegram Digest",
"type": "n8n-nodes-base.code",
"position": [
410,
500
],
"parameters": {
"jsCode": "// Get the structured output from AI Agent\nconst agentOutput = $('AI Digest Summarizer').first().json.output;\nconst imageData = $input.first().json.data;\n\n// Build Telegram message with emojis\nlet message = '\ud83e\udd16 *Daily AI News Digest*\\n';\nmessage += '\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\\n\\n';\n\nif (agentOutput && agentOutput.stories) {\n  agentOutput.stories.forEach((story, index) => {\n    const emoji = ['\ud83d\udd39', '\ud83d\udd38', '\ud83d\udca1', '\ud83d\ude80', '\u26a1'][index] || '\ud83d\udcf0';\n    message += `${emoji} *${story.title}*\\n`;\n    message += `${story.summary}\\n`;\n    message += `\ud83d\udd17 [Read more](${story.link})\\n\\n`;\n  });\n}\n\nmessage += '\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\u2501\\n';\nmessage += `\ud83d\udcc5 ${new Date().toLocaleDateString('ja-JP', { year: 'numeric', month: 'long', day: 'numeric' })}`;\n\nreturn [\n  {\n    json: {\n      telegramMessage: message,\n      imageUrl: imageData && imageData[0] ? imageData[0].url : null\n    }\n  }\n];"
},
"typeVersion": 2
},
{
"id": "5582fe93-f675-4350-99c4-5a22081ed5cf",
"name": "Dispatch Telegram Post",
"type": "n8n-nodes-base.telegram",
"position": [
640,
490
],
"webhookId": "cfa90eef-c5a9-4410-8ba7-28ccba3a75af",
"parameters": {
"file": {
"__rl": true,
"mode": "url",
"value": "={{ $json.imageUrl }}"
},
"chatId": {
"__rl": true,
"mode": "id",
"value": "={{ $('Setup Variables').first().json.telegramChatId }}"
},
"operation": "sendPhoto",
"additionalFields": {
"caption": "={{ $json.telegramMessage }}",
"parse_mode": "Markdown"
}
},
"credentials": {
"telegramApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1.2
},
{
"id": "45b1b8d2-fc34-4198-992e-157e6405690f",
"name": "Conversation AI Agent",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
570,
710
],
"parameters": {
"text": "You are a helpful AI assistant that can discuss the latest AI news. Answer questions about AI developments, trends, and stories. Be informative and engaging.",
"options": [],
"promptType": "define"
},
"typeVersion": 2.2
},
{
"id": "227555e4-2f69-4ef6-8fe6-ac8ba4bcd9e5",
"name": "Chat Model Engine",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
560,
900
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "gpt-4.1-mini"
},
"options": []
},
"credentials": {
"openAiApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1.2
},
{
"id": "e61ad8d7-5780-4c44-b8d1-4ed02922d3d3",
"name": "Chat Window Memory",
"type": "@n8n/n8n-nodes-langchain.memoryBufferWindow",
"position": [
760,
900
],
"parameters": {
"contextWindowLength": 40
},
"typeVersion": 1.3
},
{
"id": "16a3c03b-7dad-48fe-8448-9044ca171655",
"name": "Telegram Incoming Trigger",
"type": "n8n-nodes-base.telegramTrigger",
"position": [
780,
700
],
"webhookId": "543637b1-b31e-4b4a-90fd-f4f3324fb36a",
"parameters": {
"updates": [
"message"
],
"additionalFields": []
},
"credentials": {
"telegramApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1.2
}
],
"active": false,
"pinData": [],
"settings": {
"callerPolicy": "workflowsFromSameOwner",
"availableInMCP": false,
"executionOrder": "v1"
},
"versionId": null,
"connections": {
"Telegram Incoming Trigger": {
"main": [
[
{
"node": "Conversation AI Agent",
"type": "main",
"index": 0
}
]
]
},
"Create AI Cover Art": {
"main": [
[
{
"node": "Compose Telegram Digest",
"type": "main",
"index": 0
}
]
]
},
"OpenAI Chat Engine": {
"ai_languageModel": [
[
{
"node": "AI Digest Summarizer",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Scheduled Morning Trigger": {
"main": [
[
{
"node": "Setup Variables",
"type": "main",
"index": 0
}
]
]
},
"Chat Window Memory": {
"ai_memory": [
[
{
"node": "Conversation AI Agent",
"type": "ai_memory",
"index": 0
}
]
]
},
"Retrieve News Feed A": {
"main": [
[
{
"node": "Combine & Rank Stories",
"type": "main",
"index": 0
}
]
]
},
"Retrieve News Feed B": {
"main": [
[
{
"node": "Combine & Rank Stories",
"type": "main",
"index": 0
}
]
]
},
"Retrieve News Feed C": {
"main": [
[
{
"node": "Combine & Rank Stories",
"type": "main",
"index": 0
}
]
]
},
"Setup Variables": {
"main": [
[
{
"node": "Retrieve News Feed A",
"type": "main",
"index": 0
},
{
"node": "Retrieve News Feed B",
"type": "main",
"index": 0
},
{
"node": "Retrieve News Feed C",
"type": "main",
"index": 0
}
]
]
},
"Combine & Rank Stories": {
"main": [
[
{
"node": "AI Digest Summarizer",
"type": "main",
"index": 0
}
]
]
},
"Compose Telegram Digest": {
"main": [
[
{
"node": "Dispatch Telegram Post",
"type": "main",
"index": 0
}
]
]
},
"AI Digest Summarizer": {
"main": [
[
{
"node": "Create AI Cover Art",
"type": "main",
"index": 0
}
]
]
},
"Structured Result Parser": {
"ai_outputParser": [
[
{
"node": "AI Digest Summarizer",
"type": "ai_outputParser",
"index": 0
}
]
]
},
"Chat Model Engine": {
"ai_languageModel": [
[
{
"node": "Conversation AI Agent",
"type": "ai_languageModel",
"index": 0
}
]
]
}
}
}