How to Repurpose One Blog Post Into LinkedIn and X With n8n

Updated 18 October 2026 · n8n, automation, social media, content repurposing

Here is how to repurpose one blog post into LinkedIn and X posts with n8n. A trigger picks up the new post and a node pulls out the article text. An AI model then writes a separate draft for each network, a check catches length and formatting problems, and a publishing node posts through each platform's OAuth credentials.

That pattern is easy to find. As of 18 October 2026, n8n's own template library has several versions of it built on OpenAI or Gemini. The templates leave out the part a freelancer has to live with: the monthly bill, the steps that fail without warning, and the moments when a person has to step in. This guide covers the build first. Then it spends most of its time on those gaps.

What the workflow does, in plain terms

The workflow turns one finished article into drafts written for each platform, and it can publish them if you let it. The templates on n8n.io, checked on 18 October 2026, start from an RSS feed, a URL sent to a webhook, or a manual run. Some of them add a schedule trigger as well. From there they extract the article content, send it to OpenAI or Gemini, and publish to LinkedIn and X through OAuth credentials.

Two details in those templates are worth copying. The blog-to-social template for LinkedIn, X and Reddit stores RSS identifiers in a PostgreSQL database, so each article is processed only once. One template also generates image prompts or images with OpenAI. That step is optional, and you can leave it out of the core text workflow.

If you are still deciding whether n8n is the right tool for jobs like this, our piece on whether n8n is worth learning for solo automation freelancers covers the bigger question. This article assumes you have already decided to build.

The parts you need before you start

You need an n8n instance, an AI model API key, and a developer account with OAuth credentials for each social network. Here is how the pieces fit, with the weak point of each one.

The developer accounts take the longest. Both LinkedIn and X make you register an app and authorize it before n8n can post for you. Plan an afternoon for this alone.

How to repurpose one blog post into LinkedIn and X posts with n8n, step by step

The build follows the table from top to bottom, and each step below says what to set and what to test. Build it in drafts-only mode first. Add publishing last.

Pick a trigger that matches how you publish

Use an RSS trigger if your blog has a feed and you publish on a regular rhythm. Use a webhook if your CMS can call a URL when a post goes live. Use a manual trigger with a URL field while you are testing, because it lets you rerun the same article as often as you like without touching the feed.

Extract the article text, and only the article text

Fetch the page with an HTTP Request node, then pull out the main content with an HTML extraction step that targets your theme's article container. Open the output and read it before you send anything to a model. If you see "Subscribe to our newsletter" or a cookie notice in the text, your selector is too broad. The model will treat that junk as content and quote it.

Write separate prompts for LinkedIn and X

Give each network its own prompt, because the formats pull in different directions. A LinkedIn prompt can ask for a short opening line, a few short paragraphs and one question at the end. An X prompt should ask for a single tight post, or a short thread if you want one, with the main point up front.

Put one rule in both prompts in plain words: use only facts, numbers and names that appear in the article text provided. The templates checked on 18 October 2026 make no promise about factual accuracy, originality or tone, and a model asked to "make it engaging" will happily invent a statistic. Ask the model to return structured output with separate fields for each draft, so the next nodes can check them one by one.

Check length and format before anything posts

Add a check node after the AI step that counts characters in the X draft and routes anything too long back for a rewrite or into a "needs editing" branch. n8n's X and LinkedIn content generator template, as published in October 2026, includes a character-limit check of this kind. Look up the current limits on each platform's help pages before you hard-code a number, and write down the date you checked. Also strip any Markdown the model adds, since asterisks and pound signs show up as literal characters in a LinkedIn post.

Add a human approval gate

Send both drafts to wherever you already look every day: email, Slack or a row in a spreadsheet. Publish only after you approve. You can do this inside one workflow with a wait-for-approval step, or you can split it into a drafting workflow and a publishing workflow. The split is easier to debug. It also changes your execution count, which matters for cost (see below).

Publish and record what went out

Connect the LinkedIn and X nodes with your OAuth credentials. If you post images to X, check whether your setup needs both OAuth2 and OAuth1, as one n8n template states it does for media uploads. After each post, write the article identifier, the post URLs and the date to your storage table. The next run checks that table first and skips anything already done.

Our own AI Content Repurposer Kit for n8n is a pre-built version of this pattern, listed at $19 on its Payhip page as of October 2026. It is our product, so weigh that. The free templates on n8n.io get you to the same place with more setup time.

What repurposing one blog post with n8n costs per month

The monthly cost is the sum of several separate bills, and no source checked for this article adds them up. The top results all explain the build and none of them prices the whole thing. Here is every line item, so you can fill in your own numbers.

1. n8n gets cheaper per step as the chain grows. n8n's pricing page, checked on 18 October 2026, says hosted plans are billed per complete workflow execution, however many steps that run contains. The research for this article could not confirm current plan prices, so none are quoted here. Our breakdown of n8n Cloud pricing against self-hosting walks through the plan logic. A third-party pricing guide from CloudZero, read on 18 October 2026, describes a free self-managed Community Edition. We have not verified its current limits, and you still pay for the server it runs on.
2. AI model costs scale with your article length. OpenAI and Gemini bill by usage, and the templates do not say what a run costs. Your spend grows with article length, the number of drafts you ask for, and every rewrite loop from the length check.
3. X API access is the line to price first. The templates require an X developer account and OAuth credentials, and they do not say what that access costs. Read X's developer pricing before you build anything, because this is the item most likely to change your plan.
4. LinkedIn API access needs its own app. You need a LinkedIn developer app, and the templates do not state any cost for it. Check the terms for posting to a personal profile compared with a company page.
5. Deduplication storage can reuse what you have. The PostgreSQL approach needs a Postgres instance. If you self-host n8n, you may already run one.
6. Image generation stays optional. Leave it out until the text workflow runs cleanly, since it adds a second model bill.
7. Your review time is a real cost. Approval takes minutes per post, and you should count it.

To estimate the n8n line, count executions per article. A single workflow that drafts, waits for approval and publishes is one run per article plus any retries. A split setup is at least one run for drafting and one for publishing. Check whether empty trigger checks show up in your execution log. If they do, they count as well.

Where the workflow breaks and needs a human

The weak points are extraction, AI output, credentials and duplicates, in that order. n8n's blog-to-social template itself, checked on 18 October 2026, tells users to test article extraction, the AI output, and whether posts appear correctly on each account. That warning is in the template for a reason. No source checked for this article publishes failure rates, so treat each stage as something to watch.

Extraction breaks when you change your theme, add a paywall, or publish a post with an unusual layout. The fix is a simple guard. If the extracted text is much shorter than usual or contains your newsletter boilerplate, stop the run and alert yourself.

AI drafts break more subtly. A draft can be the right length and still misstate your argument, invent a figure, or sound like a press release. The approval gate exists for this, and it is the step people remove first once they get comfortable.

Credentials break on someone else's schedule. OAuth connections can be revoked or expire, and the publishing node then fails. Set up an error workflow in n8n that emails you when a run fails, so a dead connection doesn't hide for a week.

Duplicates come from feeds that re-list edited posts and from manual reruns during testing. Storing the article identifier and checking it at the start of every run, as the PostgreSQL template does, prevents this.

Pros and cons of building this in n8n

n8n gives you full control of prompts and flow, and you pay for that control in setup and maintenance.

Pros:

• You control the prompts for each network, so LinkedIn and X drafts can follow your own voice rules.
• Long chains stay cheap on hosted plans, since n8n's pricing page in October 2026 bills per full execution and not per step.
• You can self-host, which keeps article drafts on your own server.
• You start from working templates in n8n's library instead of a blank canvas.

Cons:

• Developer accounts take real effort, because both networks need an app, OAuth setup and authorization before anything posts.
• X media uploads add complexity, since one template needs two kinds of X credentials for images.
• Accuracy is still on you, because none of the templates promise correct facts, originality or suitable tone.
• The full cost is scattered, with API and model prices sitting on pages you have to check yourself.
• Maintenance never ends, since theme changes, API changes and expired credentials all break runs.
• There is no independent reliability data, so you are the test case.

Who should not build an n8n blog repurposing workflow

Skip this build if you publish only now and then. Pasting the article into a chat model and editing the result takes less time than registering two developer apps. Skip it if you will not review drafts, because unreviewed AI posts on your own accounts are a risk to your reputation that no workflow fixes.

Skip it too if what you really want is a content calendar with analytics. A scheduler does that job better, and our roundup of AI tools for social media scheduling compares them. Agencies posting to client accounts should think twice unless each client has signed off on an approval process.

FAQ

Can n8n post to LinkedIn and X automatically?

Yes. Templates in n8n's library, checked on 18 October 2026, publish to both through OAuth credentials once you have developer accounts set up. The template documentation still tells you to test whether posts appear correctly on each account. Run it in drafts-only mode first.

Do I have to pay for n8n to repurpose blog posts?

You pay either for hosting or for a hosted plan. n8n's pricing page, checked on 18 October 2026, bills hosted plans per complete workflow execution. A third-party guide from CloudZero describes a free self-managed Community Edition, though you still pay for the server and should check its current terms.

Should I use OpenAI or Gemini for the drafts?

The templates support both, so start with whichever API you already have access to. Run the same three or four articles through each and compare tone and accuracy. Keep the prompt identical between the two so the comparison is fair.

How do I stop the same blog post being shared twice?

Store each article's identifier after you process it and check that list at the start of every run. One n8n template does this with PostgreSQL, and a spreadsheet works for low volume. Clear test entries before you go live.

Is it safe to let the workflow publish without review?

It is risky, and the templates give you no reason to trust it. They make no promise about factual accuracy or tone, and they tell you to test the AI output yourself. Keep a human approval step at least until you have read many drafts without needing a fix.

Verdict

Build it if you publish on a steady rhythm, already run n8n, and don't mind the developer account setup. Start with a manual trigger and drafts sent to your inbox, add the approval gate, and switch on publishing only when the drafts stop needing edits. If you post rarely or will not review drafts, a chat model and copy-paste is the cheaper and safer choice.

Related reading

• n8n vs Zapier for small business automation
• How to sort Gmail automatically with n8n and Claude
• Make.com vs n8n: which automation tool fits

About the author

The Tool Signal team writes about AI and automation tools for freelancers and small businesses, with a focus on real prices and real limits. We did not run this exact workflow for this article. It is built from n8n's published templates and pricing page as they stood on 18 October 2026, and every product claim is listed below with its source and status. Read that list before you trust any guide on how to repurpose one blog post into LinkedIn and X posts with n8n.