January 22, 2026

Gemini + LinkedIn: carousels published consistently

Lisa Granqvist Partner Workflow Automation Expert

Get a free AI assessment → ⬇️ Use template

Posting LinkedIn carousels sounds simple until you try doing it consistently. You find a story, rewrite it, format slides, export a PDF, upload it, then craft a caption. Next day, repeat. It’s not hard work. It’s the never-ending kind.

This hits marketing managers first, because the content calendar never stops. But agency owners trying to keep multiple clients visible feel it too, and solo founders get stuck in the same loop. This Gemini LinkedIn carousels automation turns a TechRadar story into a ready-to-post carousel PDF and publishes it for you, so you can keep a steady cadence without living in Canva.

Below you’ll see how the workflow runs at 6:00 AM, how it chooses a story that fits your profile, and how PostNitro plus LinkedIn upload APIs handle the boring parts.

How This Automation Works

See how this solves the problem:

n8n Workflow Template: Gemini + LinkedIn: carousels published consistently

Click to explore

flowchart LR

    subgraph sg0["6:00 AM Flow"]
        direction LR
        n0@{ icon: "mdi:play-circle", form: "rounded", label: "6:00 AM Trigger", pos: "b", h: 48 }
        n1["<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/form.svg' width='40' height='40' /></div><br/>On form submission"]
        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/>TechRadar News"]
        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/>Parse HTML"]
        n4@{ icon: "mdi:brain", form: "rounded", label: "Gemini Flash 2.5", 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/>Parse LLM Response"]
        n6@{ icon: "mdi:cog", form: "rounded", label: "ImportSlides embedPost", pos: "b", h: 48 }
        n7@{ icon: "mdi:robot", form: "rounded", label: "Decide which News to Choose", pos: "b", h: 48 }
        n8["<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/>Linkedin User URN"]
        n9["<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/>Get Upload URL"]
        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/httprequest.dark.svg' width='40' height='40' /></div><br/>Upload PDF"]
        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/httprequest.dark.svg' width='40' height='40' /></div><br/>Download PDF"]
        n12["<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/>Post to LinkedIn"]
        n13@{ icon: "mdi:robot", form: "rounded", label: "Gemini Generate", pos: "b", h: 48 }
        n14@{ icon: "mdi:robot", form: "rounded", label: "Gemini Generate1", 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/code.svg' width='40' height='40' /></div><br/>Parse LLM Response1"]
        n16["<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/>Get Title and Description"]
        n3 --> n7
        n10 --> n14
        n11 --> n10
        n9 --> n11
        n2 --> n3
        n0 --> n2
        n13 --> n15
        n4 -.-> n7
        n14 --> n12
        n8 --> n9
        n1 --> n16
        n5 --> n16
        n15 --> n6
        n6 --> n8
        n16 --> n13
        n7 --> n5
    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,n1 trigger
    class n7,n13,n14 ai
    class n4 aiModel
    class n2,n8,n9,n10,n11,n12 api
    class n3,n5,n15,n16 code
    classDef customIcon fill:none,stroke:none
    class n1,n2,n3,n5,n8,n9,n10,n11,n12,n15,n16 customIcon

The Challenge: Posting carousels without losing your mornings

Carousels are great for reach, saves, and “I needed this” comments. The downside is the production line behind each post. You’re copying headlines from a feed, deciding what’s relevant, rewriting it into a narrative, then converting that narrative into slide-friendly chunks. After that you still have the export-and-upload dance, plus a caption that doesn’t sound like it was stitched together at the last minute. Miss two days and your “we post consistently” plan quietly dies.

It adds up fast. And the friction compounds in the same spots every time:

Picking a daily story becomes a mini research project, especially when you’re trying to match a niche and not just repost whatever is trending.
Carousel formatting is tedious, because slide structure needs clarity, not long paragraphs, and you end up rewriting the same idea three times.
Uploading PDFs to LinkedIn is annoyingly manual, and doing it on a phone makes it worse.
When someone on the team “just handles posting,” quality slips or approvals get skipped, which means the brand voice drifts over time.

The Fix: TechRadar → Gemini → PostNitro → LinkedIn, fully automated

This workflow runs every morning at 6:00 AM (or whenever you choose) and pulls the latest TechRadar news feed. It parses the RSS items into clean JSON, then Gemini evaluates which story best fits your profile and audience. Once a headline is selected, the workflow generates carousel-ready copy that’s already structured for slides, not blog paragraphs. PostNitro turns that structured content into a designed carousel and returns a PDF file. Finally, the workflow uploads the PDF to LinkedIn, generates post text with Gemini, and publishes the post using LinkedIn’s upload flow and post API calls. You can also trigger the whole thing with a form submission when you want to override the daily pick with your own title and description.

The workflow starts from a schedule trigger or a form trigger. Gemini handles both selection and writing, while PostNitro handles the visual build and PDF output. LinkedIn upload steps grab the file ID and attach it to the final post so publishing happens in one pass.

What Changes: Before vs. After

What This Eliminates

Impact You’ll See

Manually scanning TechRadar and copying a story into your notes.
Rewriting the same article into “slide-sized” chunks by hand.
Exporting a carousel PDF and doing repeated uploads and retries.
Starting from a blank caption every time you publish.

Most teams get about 2 hours back per carousel day.
More consistent posting, because the workflow runs on a schedule even when your calendar is packed.
Fewer “formatting mistakes” like overcrowded slides or uneven pacing across pages.
A repeatable content standard, so your voice stays coherent across weeks.
Easier scaling for multiple brands, since the same workflow can be duplicated and tuned.

Real-World Impact

Say you publish 5 carousels a week. Manually, a typical run looks like 20 minutes to pick and summarize a story, about 45 minutes rewriting it into slides, another 20 minutes creating/exporting a PDF, plus 10 minutes uploading and writing a caption. That’s roughly 1.5 hours per post, or about 7 hours a week. With this automation, you spend maybe 10 minutes reviewing the generated carousel copy (or submitting a form with your own angle), then it builds the PDF and publishes while you’re doing other work.

Requirements

n8n instance (try n8n Cloud free)
Self-hosting option if you prefer (Hostinger works well)
Google Gemini to select stories and write carousels.
PostNitro to render carousel slides into a PDF.
LinkedIn API access (get it from your LinkedIn developer app settings)

Skill level: Intermediate. You’ll be connecting APIs and pasting credentials, but you won’t be writing an application.

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

The Workflow Flow

A scheduled or manual trigger starts the run. At 6:00 AM the schedule fires automatically, or you can submit a form when you want to feed a specific title and description into the pipeline.

The workflow fetches TechRadar and cleans the feed. An HTTP request pulls the latest items, then a small code step converts the RSS output into readable JSON so the AI can evaluate it reliably.

Gemini selects the best story and writes slide-ready content. One Gemini step ranks which news item “fits” your profile, then another Gemini step generates the carousel copy as structured data that can be split into slides. A parsing step checks the response so PostNitro receives clean fields.

PostNitro renders the carousel, then LinkedIn publishes it. PostNitro returns a PDF file, the workflow initializes an upload session, uploads the PDF, gets the document ID, generates the post text, and publishes the final LinkedIn post through HTTP requests.

You can easily modify the source feed (TechRadar) to another RSS source based on your needs. See the full implementation guide below for customization options.

Step-by-Step Implementation Guide

Step 1: Configure the Trigger Type

Set up the scheduled and manual triggers that start the workflow, so it can run daily or on-demand via form submission.

Open Scheduled Morning Trigger and set the schedule rule to run at 6 (triggerAtHour) for your daily automation.
Open Form Submission Trigger and confirm the Path is create-carousal and Button Label is Create Carousal.
In Form Submission Trigger, keep Form Title as Linkedin Carousal and Response Mode as lastNode.
Ensure the form fields include Title and Description with the provided placeholders so manual submissions can route into content generation.

Step 2: Connect the News Source and Parse RSS

Fetch the RSS feed and parse items into structured JSON for AI selection.

Configure Fetch Tech News Feed with URL set to https://www.techradar.com/feeds.xml.
In Extract RSS Items, keep the provided JavaScript that parses the RSS XML and returns a result array.
Verify the connection flow: Scheduled Morning Trigger → Fetch Tech News Feed → Extract RSS Items.

Step 3: Set Up AI Headline Selection and Summary Assembly

Use Gemini to choose the best headline, then merge the title and description into a unified summary for downstream AI steps.

Open Gemini Flash Selector and select your Gemini model connection. Credential Required: Connect your googlePalmApi credentials.
In Select Relevant Headline, set Text to the full prompt expression beginning with =You are an AI assistant... and ensure it references the RSS list using {{ JSON.stringify($json.result.map((item, index) => ({ number: index+1, title: item.title })), null, 2) }}.
Confirm Gemini Flash Selector is connected as the language model for Select Relevant Headline (credentials are added on Gemini Flash Selector, not the chain node).
In Read LLM Choice, keep the code that parses the selected number and reads from Extract RSS Items: const item = $('Extract RSS Items').first().json.result[number].
In Assemble Title Summary, keep Mode as runOnceForEachItem and the code that returns combinedTitle and combinedDescription.

Tip: The workflow accepts both automated RSS input and manual form input, because Form Submission Trigger and Read LLM Choice both output to Assemble Title Summary.

Step 4: Generate Carousel Slides and Build the PDF

Create the carousel content with Gemini, parse the JSON into slides, and import into PostNitro to produce a PDF.

In Generate Carousel Copy, choose model models/gemini-2.5-flash and keep the long prompt that outputs valid JSON only. Credential Required: Connect your googlePalmApi credentials.
Ensure the prompt includes the expressions for input: {{ $json.combinedTitle }} and {{ $json.combinedDescription }}.
In Parse Carousel JSON, keep the code that removes code fences and parses JSON from $input.first().json.content.parts[0].text.
Configure PostNitro Slide Import with Operation set to importSlides, Slides JSON set to ={{ JSON.stringify( $json.slides ) }}, and replace [YOUR_ID] for both brandId and templateId. Credential Required: Connect your postNitroApi credentials.

⚠️ Common Pitfall: If Generate Carousel Copy returns anything other than valid JSON, Parse Carousel JSON will fail. Keep the prompt’s “Return only raw JSON” requirement intact.

Step 5: Configure LinkedIn Upload and Publishing

Upload the PostNitro PDF to LinkedIn, generate the post commentary, and publish the carousel post.

In Retrieve LinkedIn User, set URL to https://api.linkedin.com/v2/userinfo and use Authentication as predefined credential type. Credential Required: Connect your linkedInOAuth2Api credentials.
Configure Initialize Upload Link with URL https://api.linkedin.com/rest/documents?action=initializeUpload, Method POST, and JSON Body as ={"initializeUploadRequest":{"owner":"urn:li:person:{{ $json.sub }}"}}. Keep header LinkedIn-Version: 202307. Credential Required: Connect your linkedInOAuth2Api credentials.
Set Fetch PDF File URL to ={{ $('PostNitro Slide Import').item.json.data.result.data }} and output the file as =my.pdf in the response settings.
In Send PDF Upload, set URL to ={{ $json.value.uploadUrl }}, Method to PUT, Content Type to binaryData, and Input Data Field Name to =my.pdf. Include header Content-Type: application/pdf. Credential Required: Connect your linkedInOAuth2Api credentials.
In Generate Post Text, choose model models/gemini-2.5-flash and keep the prompt that outputs raw text using {{ $('Assemble Title Summary').item.json.combinedTitle }} and {{ $('Assemble Title Summary').item.json.combinedDescription }}. Credential Required: Connect your googlePalmApi credentials.
In Publish LinkedIn Post, set URL to https://api.linkedin.com/rest/posts, Method POST, and keep the JSON body with expressions: {{ $('Retrieve LinkedIn User').item.json.sub }}, {{ JSON.stringify($json.content.parts[0].text) }}, {{ $('Parse Carousel JSON').item.json.slides[0].heading }}, and {{ $('Initialize Upload Link').item.json.value.document }}. Keep header LinkedIn-Version: 202408. Credential Required: Connect your linkedInOAuth2Api credentials.

Step 6: Test and Activate Your Workflow

Run a test execution to verify the full chain from content generation to LinkedIn publishing.

Click Execute Workflow and use either Form Submission Trigger with sample Title/Description or wait for Scheduled Morning Trigger to run.
Confirm the flow executes in order: Fetch Tech News Feed → Extract RSS Items → Select Relevant Headline → Read LLM Choice → Assemble Title Summary → Generate Carousel Copy → Parse Carousel JSON → PostNitro Slide Import → Retrieve LinkedIn User → Initialize Upload Link → Fetch PDF File → Send PDF Upload → Generate Post Text → Publish LinkedIn Post.
Successful execution should end with a LinkedIn post created and a valid document ID returned in Publish LinkedIn Post.
Toggle the workflow to Active to enable daily publishing at 6 AM and on-demand submissions.

🔒

Unlock Full Step-by-Step Guide

Get the complete implementation guide + downloadable template

Watch Out For

LinkedIn API credentials can expire or require extra permissions. If things break, check your LinkedIn developer app settings and token validity 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 Gemini LinkedIn carousels automation?

About an hour if you already have the API keys ready.

Can non-technical teams implement this carousel publishing?

Yes, but someone needs to be comfortable with API keys and testing a few runs. No “real coding,” just careful setup and a bit of troubleshooting.

Is n8n free to use for this Gemini LinkedIn carousels 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 Gemini API usage and your PostNitro plan.

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 Gemini LinkedIn carousels solution to my specific challenges?

You can swap the TechRadar fetch step for another RSS feed by changing the “Fetch Tech News Feed” HTTP request. If you want stricter selection, tune the prompt in the “Gemini Flash Selector” so it only picks stories in your niche. Common customizations include changing the number of slides in the generated JSON, adjusting PostNitro template and brand IDs, and routing drafts to Google Sheets before publishing.

Why is my LinkedIn connection failing in this workflow?

Usually it’s expired tokens or missing LinkedIn permissions for uploads and posting. Regenerate your access token, confirm the app has the correct scopes, then update credentials in n8n and re-run one test item. If the PDF upload succeeds but publishing fails, check the “Initialize Upload Link” and “Publish LinkedIn Post” HTTP responses for a useful error message. Rate limiting can also show up if you’re running many tests in a short burst.

What’s the capacity of this Gemini LinkedIn carousels solution?

On n8n Cloud Starter, you can run a few thousand executions per month, which is plenty for daily posting. If you self-host, there’s no execution cap, but it depends on your server and how long PostNitro rendering takes. Practically, most teams run this once per day per brand, plus a few manual form-trigger runs for special posts.

Is this Gemini LinkedIn carousels automation better than using Zapier or Make?

Often, yes. This flow has multiple stages (RSS parsing, AI selection, structured JSON generation, PDF handling, multi-call LinkedIn upload), and n8n handles that kind of branching and data shaping without turning into a fragile pile of zaps. The self-hosted option matters too, because community nodes (like PostNitro’s) are common in workflows like this. Zapier or Make can still work if you simplify the process, like generating text only and uploading manually. If you want the exact “PDF carousel to LinkedIn” path and you’re unsure which route is fastest, Talk to an automation expert.

Once this is live, your LinkedIn carousel pipeline runs in the background and your mornings stay yours. Review when you want, publish consistently, and move on to the work that actually grows the business.

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": "5d4e4724ad839e9c5d10f5c4d0649ec0383e2d4a6b6cbf2b3c9e48244a679149",
"templateCredsSetupCompleted": null
},
"name": "Automated LinkedIn Carousel Publisher",
"tags": [],
"nodes": [
{
"id": "flowpast-topbar-7733",
"name": "Flowpast Branding",
"type": "n8n-nodes-base.stickyNote",
"position": [
0,
20
],
"parameters": {
"color": 7,
"width": 1235,
"height": 80,
"content": "## Flowpast.com | Automation Workflow Library\n**\ud83d\udcd6 Full tutorial & setup guide:** flowpast.com"
},
"typeVersion": 1
},
{
"id": "5f5e3dce-7b82-4654-8cc4-088cfd799a36",
"name": "Scheduled Morning Trigger",
"type": "n8n-nodes-base.scheduleTrigger",
"position": [
50,
185
],
"parameters": {
"rule": {
"interval": [
{
"triggerAtHour": 6
}
]
}
},
"typeVersion": 1.2
},
{
"id": "8fa6d50b-6c31-423a-b233-38f8585e6905",
"name": "Form Submission Trigger",
"type": "n8n-nodes-base.formTrigger",
"position": [
1045,
400
],
"webhookId": "181a4bfd-c5ac-4edb-93e0-34dd66789456",
"parameters": {
"options": {
"path": "create-carousal",
"ignoreBots": true,
"buttonLabel": "Create Carousal"
},
"formTitle": "Linkedin Carousal",
"formFields": {
"values": [
{
"fieldLabel": "Title",
"placeholder": "Nvidia released a new GPU rtx 4090 that supports state of the art gaming and crypto. The really pushed the limits with it.",
"requiredField": true
},
{
"fieldType": "textarea",
"fieldLabel": "Description",
"placeholder": "Nvidia released a new GPU rtx 4090 that supports state of the art gaming and crypto. The really pushe ....",
"requiredField": true
}
]
},
"responseMode": "lastNode",
"formDescription": "Please fill out the form below"
},
"typeVersion": 2.2
},
{
"id": "2224d22d-62ac-4bfa-94e1-260500addc52",
"name": "Fetch Tech News Feed",
"type": "n8n-nodes-base.httpRequest",
"position": [
250,
235
],
"parameters": {
"url": "https://www.techradar.com/feeds.xml",
"options": []
},
"typeVersion": 4.2
},
{
"id": "11057068-72af-4a85-bdd0-1094be3109ce",
"name": "Extract RSS Items",
"type": "n8n-nodes-base.code",
"position": [
410,
250
],
"parameters": {
"jsCode": "function parseRssFeed(jsonData) {\n  try {\n    // Parse JSON input\n    const dataObj = JSON.parse(jsonData);\n    const xmlString = dataObj[0].json.data;\n\n    // Simple regex-based XML parser for RSS items\n    const itemRegex = /<item>([\\s\\S]*?)<\\/item>/g;\n    const items = [];\n    let match;\n\n    // Extract all <item> tags\n    while ((match = itemRegex.exec(xmlString)) !== null) {\n      const itemContent = match[1];\n      const item = {};\n\n      // Define fields to extract\n      const fields = [\n        { tag: 'title', key: 'title' },\n        { tag: 'description', key: 'description' },\n        { tag: 'link', key: 'link' },\n        { tag: 'guid', key: 'guid' },\n        { tag: 'pubDate', key: 'pubDate' },\n        { tag: 'category', key: 'category' },\n        { tag: 'dc:creator', key: 'dc_creator' },\n        { tag: 'media:content', key: 'media_content', subtag: 'media:credit' },\n        { tag: 'media:content', key: 'media_text', subtag: 'media:text' },\n        { tag: 'media:content', key: 'media_title', subtag: 'media:title' },\n        { tag: 'dc:content', key: 'content' }\n      ];\n\n      // Extract each field\n      fields.forEach(field => {\n        let regex;\n        if (field.subtag) {\n          regex = new RegExp(`<${field.tag}[^>]*>[\\\\s\\\\S]*?<${field.subtag}[^>]*><!\\\\[CDATA\\\\[(.*?)\\\\]\\\\]></${field.subtag}>`, 'i');\n        } else {\n          regex = new RegExp(`<${field.tag}(?:[^>]*)><!\\\\[CDATA\\\\[(.*?)\\\\]\\\\]></${field.tag}>|<${field.tag}(?:[^>]*)>(.*?)</${field.tag}>`, 'i');\n        }\n        \n        const fieldMatch = itemContent.match(regex);\n        if (fieldMatch) {\n          let value = fieldMatch[1] || fieldMatch[2] || '';\n          // Remove HTML tags\n          value = value.replace(/<[^>]+>/g, '').replace(/ /g, ' ').replace(/&/g, '&').trim();\n          if (value) {\n            item[field.key] = value;\n          }\n        }\n      });\n\n      // Handle multiple categories\n      const categoryMatches = itemContent.match(/<category><!\\[CDATA\\[(.*?)\\]\\]><\\/category>/g);\n      if (categoryMatches) {\n        item.category = categoryMatches\n          .map(cat => cat.match(/<category><!\\[CDATA\\[(.*?)\\]\\]><\\/category>/)[1])\n          .join(', ');\n      }\n\n      if (Object.keys(item).length > 0) {\n        items.push(item);\n      }\n    }\n\n    return items;\n  } catch (error) {\n    throw new Error(`Error processing RSS feed: ${error?.message}`);\n  }\n}\n\n// n8n Function node implementation\ntry {\n  const jsonData = JSON.stringify($input.all());\n  const result = parseRssFeed(jsonData);\n  return {result, error: false, message: null}\n} catch (error) {\n  console.log(\"Error : \",error)\n  return {result: [], error: true, message: error?.message}\n}"
},
"typeVersion": 2
},
{
"id": "6581b0bb-373e-4986-9951-79957bfb482f",
"name": "Gemini Flash Selector",
"type": "@n8n/n8n-nodes-langchain.lmChatGoogleGemini",
"position": [
705,
120
],
"parameters": {
"options": []
},
"credentials": {
"googlePalmApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1
},
{
"id": "d30affe8-ec6e-4b82-859a-9b306222059b",
"name": "Select Relevant Headline",
"type": "@n8n/n8n-nodes-langchain.chainLlm",
"position": [
650,
245
],
"parameters": {
"text": "=You are an AI assistant. Your only task is to select the single most relevant news item from the provided news title list.\n\nRelevance criteria (in order of importance):\n1. n8n Automation\n2. AI and Technology News\n3. Personal brand development and motivation\n\nExclusion rules:\n- Ignore any news about love, relationships, entertainment gossip, or censored/inappropriate topics.\n\nNews list (each item has a number and a title):\n```\n[\n{{ JSON.stringify($json.result.map((item, index) => ({ number: index+1, title: item.title })), null, 2) }}\n]\n```\n\nReturn format:\n- Return only the number of the most relevant news item.\n- Do not return text, explanations, or anything else. Just a number\n\nExample valid responses:\n```\n6\n```",
"batching": [],
"promptType": "define"
},
"typeVersion": 1.7
},
{
"id": "d77b44a6-68d3-4533-b909-d0794cd61c29",
"name": "Read LLM Choice",
"type": "n8n-nodes-base.code",
"position": [
945,
195
],
"parameters": {
"jsCode": "const number = parseInt($input.first().json.text)\nconst item = $('Extract RSS Items').first().json.result[number]\n\nreturn item"
},
"typeVersion": 2
},
{
"id": "acb524f6-629a-4d8d-a9d9-aee282f5a44e",
"name": "Assemble Title Summary",
"type": "n8n-nodes-base.code",
"position": [
850,
445
],
"parameters": {
"mode": "runOnceForEachItem",
"jsCode": "const combinedTitle = $json.title || $json.Title\nconst combinedDescription = $json.content || $json.Description\n\nreturn {\n  combinedTitle,\n  combinedDescription\n}"
},
"typeVersion": 2
},
{
"id": "10af8746-8f9a-4f18-8e70-c681e35c0ee5",
"name": "Generate Carousel Copy",
"type": "@n8n/n8n-nodes-langchain.googleGemini",
"position": [
705,
370
],
"parameters": {
"modelId": {
"__rl": true,
"mode": "list",
"value": "models/gemini-2.5-flash",
"cachedResultName": "models/gemini-2.5-flash"
},
"options": [],
"messages": {
"values": [
{
"content": "=You are an AI Agent that generates LinkedIn carousel post content.\n\nIMPORTANT: Return only raw JSON as plain text. Do not call functions, do not use tool calls, do not add extra text.\n\nYour task: Write engaging, professional slide text for a LinkedIn carousel based on the provided title and description.\n\nNews Input:\n```\nTitle: {{ $json.combinedTitle }}\nDescription: {{ $json.combinedDescription }}\n```\n\nCarousel Requirements:\n- Exactly 1 `starting_slide` \u2192 strong hook, with heading, sub_heading, description, and CTA button.\n- At least 1 `body_slide` \u2192 explain key insights or takeaways from the news (you may create 2\u20133 if relevant).\n- Exactly 1 `ending_slide` \u2192 motivational close with sub_heading, heading, description, and CTA button.\n- Keep tone professional, concise, and LinkedIn-appropriate.\n- Do not include irrelevant topics (love, politics, entertainment gossip).\n- Ensure headings are short and catchy; descriptions can be 1\u20133 sentences.\n- Output must be valid JSON only, matching the schema below.\n\nResponse Format:\n```\n{\n\"slides\": [\n  {\n    \"type\": \"starting_slide\",\n    \"sub_heading\": \"string\",\n    \"heading\": \"string\",\n    \"description\": \"string\",\n    \"cta_button\": \"string\"\n  },\n  {\n    \"type\": \"body_slide\",\n    \"heading\": \"string\",\n    \"description\": \"string\"\n  },\n  {\n    \"type\": \"ending_slide\",\n    \"sub_heading\": \"string\",\n    \"heading\": \"string\",\n    \"description\": \"string\",\n    \"cta_button\": \"string\"\n  }\n]\n}\n```\n"
}
]
}
},
"credentials": {
"googlePalmApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1
},
{
"id": "a8715e77-2a98-4856-b38f-fe79b38e2199",
"name": "Parse Carousel JSON",
"type": "n8n-nodes-base.code",
"position": [
405,
450
],
"parameters": {
"jsCode": "let raw = $input.first().json.content.parts[0].text\n\n// Remove code fences if present (```json ... ```)\nraw = raw.replace(/```json\\n?/g, \"\").replace(/```/g, \"\");\n\n// Parse into JSON\nlet parsed;\ntry {\n  parsed = JSON.parse(raw);\n} catch (e) {\n  throw new Error(\"Failed to parse AI response as JSON: \" + e.message);\n}\n\n// Return slides array as n8n items\nreturn parsed"
},
"typeVersion": 2
},
{
"id": "b0399c03-d5f7-44f6-92b7-4e0a43d9de75",
"name": "PostNitro Slide Import",
"type": "@postnitro/n8n-nodes-postnitro-ai.postNitro",
"position": [
250,
410
],
"parameters": {
"brandId": "[YOUR_ID]",
"operation": "importSlides",
"slidesJson": "={{ JSON.stringify( $json.slides ) }}",
"templateId": "[YOUR_ID]"
},
"credentials": {
"postNitroApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1
},
{
"id": "3aceebb8-82f6-44b9-a574-bab608861a17",
"name": "Retrieve LinkedIn User",
"type": "n8n-nodes-base.httpRequest",
"position": [
0,
445
],
"parameters": {
"url": "https://api.linkedin.com/v2/userinfo",
"options": [],
"authentication": "predefinedCredentialType",
"nodeCredentialType": "linkedInOAuth2Api"
},
"credentials": {
"linkedInOAuth2Api": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 4.2
},
{
"id": "a150d9bd-864c-449e-a2e9-11189d4b84f8",
"name": "Initialize Upload Link",
"type": "n8n-nodes-base.httpRequest",
"position": [
155,
630
],
"parameters": {
"url": "https://api.linkedin.com/rest/documents?action=initializeUpload",
"method": "POST",
"options": [],
"jsonBody": "={\n  \"initializeUploadRequest\": {\n        \"owner\": \"urn:li:person:{{ $json.sub }}\"\n  }\n}",
"sendBody": true,
"sendHeaders": true,
"specifyBody": "json",
"authentication": "predefinedCredentialType",
"headerParameters": {
"parameters": [
{
"name": "LinkedIn-Version",
"value": "202307"
}
]
},
"nodeCredentialType": "linkedInOAuth2Api"
},
"credentials": {
"linkedInOAuth2Api": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 4.2
},
{
"id": "f2bcc3a1-e5bc-445e-b3b6-b8d594ccd4e3",
"name": "Fetch PDF File",
"type": "n8n-nodes-base.httpRequest",
"position": [
440,
595
],
"parameters": {
"url": "={{ $('PostNitro Slide Import').item.json.data.result.data }}",
"options": {
"response": {
"response": {
"responseFormat": "file",
"outputPropertyName": "=my.pdf"
}
}
}
},
"typeVersion": 4.2
},
{
"id": "01bf5969-6e00-47f4-9ae6-42c004f6fd0f",
"name": "Send PDF Upload",
"type": "n8n-nodes-base.httpRequest",
"position": [
710,
615
],
"parameters": {
"url": "={{ $json.value.uploadUrl }}",
"method": "PUT",
"options": [],
"sendBody": true,
"contentType": "binaryData",
"sendHeaders": true,
"authentication": "predefinedCredentialType",
"headerParameters": {
"parameters": [
{
"name": "Content-Type",
"value": "application/pdf"
}
]
},
"inputDataFieldName": "=my.pdf",
"nodeCredentialType": "linkedInOAuth2Api"
},
"credentials": {
"linkedInOAuth2Api": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 4.2
},
{
"id": "55937c1f-4fd0-43e2-a03d-8a78fb7410af",
"name": "Generate Post Text",
"type": "@n8n/n8n-nodes-langchain.googleGemini",
"position": [
830,
655
],
"parameters": {
"modelId": {
"__rl": true,
"mode": "list",
"value": "models/gemini-2.5-flash",
"cachedResultName": "models/gemini-2.5-flash"
},
"options": [],
"messages": {
"values": [
{
"content": "=You are an AI Agent that generates LinkedIn post description.\n\nIMPORTANT: Return only raw text.\n\nYour task: Write engaging, professional text for a LinkedIn post based on the provided title and description.\n\nNews Input:\n```\nTitle: {{ $('Assemble Title Summary').item.json.combinedTitle }}\nDescription: {{ $('Assemble Title Summary').item.json.combinedDescription }}\n```\n\nExample Response Format:\n```\n\ud83d\ude80 Just launched my first fully automated LinkedIn carousel \u2014 created end-to-end by my AI Employee (built in n8n).\n\nNo manual editing, no wasted time. The system handles content creation \u2192 carousel design \u2192 posting all on its own.\n\n\u2728 I\u2019m offering to set this up for 5 people free of cost. If you want to automate your content creation and save hours every week, drop a comment below.\n```\n"
}
]
}
},
"credentials": {
"googlePalmApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1
},
{
"id": "ddbe76f1-856d-4242-b55a-bfae1b4025c7",
"name": "Publish LinkedIn Post",
"type": "n8n-nodes-base.httpRequest",
"position": [
1135,
580
],
"parameters": {
"url": "https://api.linkedin.com/rest/posts",
"method": "POST",
"options": [],
"jsonBody": "={\n  \"author\": \"urn:li:person:{{ $('Retrieve LinkedIn User').item.json.sub }}\",\n  \"commentary\": {{ JSON.stringify($json.content.parts[0].text) }},\n  \"visibility\": \"PUBLIC\",\n  \"distribution\": {\n    \"feedDistribution\": \"MAIN_FEED\",\n    \"targetEntities\": [],\n    \"thirdPartyDistributionChannels\": []\n  },\n  \"content\": {\n    \"media\": {\n      \"title\":\"{{ $('Parse Carousel JSON').item.json.slides[0].heading }}\",\n      \"id\": \"{{ $('Initialize Upload Link').item.json.value.document }}\"\n    }\n  },\n  \"lifecycleState\": \"PUBLISHED\",\n  \"isReshareDisabledByAuthor\": false\n}",
"sendBody": true,
"sendHeaders": true,
"specifyBody": "json",
"authentication": "predefinedCredentialType",
"headerParameters": {
"parameters": [
{
"name": "LinkedIn-Version",
"value": "202408"
},
{
"name": "Content-Type",
"value": "application/json"
}
]
},
"nodeCredentialType": "linkedInOAuth2Api"
},
"credentials": {
"linkedInOAuth2Api": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 4.2
}
],
"active": false,
"pinData": [],
"settings": {
"callerPolicy": "workflowsFromSameOwner",
"errorWorkflow": "OjCzQZHTVMPt0jhc",
"executionOrder": "v1"
},
"versionId": null,
"connections": {
"Extract RSS Items": {
"main": [
[
{
"node": "Select Relevant Headline",
"type": "main",
"index": 0
}
]
]
},
"Send PDF Upload": {
"main": [
[
{
"node": "Generate Post Text",
"type": "main",
"index": 0
}
]
]
},
"Fetch PDF File": {
"main": [
[
{
"node": "Send PDF Upload",
"type": "main",
"index": 0
}
]
]
},
"Initialize Upload Link": {
"main": [
[
{
"node": "Fetch PDF File",
"type": "main",
"index": 0
}
]
]
},
"Fetch Tech News Feed": {
"main": [
[
{
"node": "Extract RSS Items",
"type": "main",
"index": 0
}
]
]
},
"Scheduled Morning Trigger": {
"main": [
[
{
"node": "Fetch Tech News Feed",
"type": "main",
"index": 0
}
]
]
},
"Generate Carousel Copy": {
"main": [
[
{
"node": "Parse Carousel JSON",
"type": "main",
"index": 0
}
]
]
},
"Gemini Flash Selector": {
"ai_languageModel": [
[
{
"node": "Select Relevant Headline",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Generate Post Text": {
"main": [
[
{
"node": "Publish LinkedIn Post",
"type": "main",
"index": 0
}
]
]
},
"Retrieve LinkedIn User": {
"main": [
[
{
"node": "Initialize Upload Link",
"type": "main",
"index": 0
}
]
]
},
"Form Submission Trigger": {
"main": [
[
{
"node": "Assemble Title Summary",
"type": "main",
"index": 0
}
]
]
},
"Read LLM Choice": {
"main": [
[
{
"node": "Assemble Title Summary",
"type": "main",
"index": 0
}
]
]
},
"Parse Carousel JSON": {
"main": [
[
{
"node": "PostNitro Slide Import",
"type": "main",
"index": 0
}
]
]
},
"PostNitro Slide Import": {
"main": [
[
{
"node": "Retrieve LinkedIn User",
"type": "main",
"index": 0
}
]
]
},
"Assemble Title Summary": {
"main": [
[
{
"node": "Generate Carousel Copy",
"type": "main",
"index": 0
}
]
]
},
"Select Relevant Headline": {
"main": [
[
{
"node": "Read LLM Choice",
"type": "main",
"index": 0
}
]
]
}
}
}