I Built An AI Agent That Ended My Content Treadmill For Good

1 article → 11 pieces of content with AI (steal it here!)

Content creation is a treadmill we all never asked to be on.

You publish an article. It gets some traction. A few likes, maybe a comment or two. Then it vanishes into the feed within hours, buried under an avalanche of new posts. The algorithm moves on. Your audience moves on. And you’re back at the keyboard, staring at a blank page, trying to conjure up something new.

This cycle repeats. Daily, if you’re disciplined. Weekly, if you’re busy. Monthly, if you’re burnt out.

The internet is relentless, with endless feeds and new creators popping up every day.

Sometimes, it feels like standing still means falling behind. And keeping up means running faster, publishing more, always feeding the machine.

Slow content slowed my growth

Solo founders, new creators, business owners building your brands—you’re already stretched thin. You have products to build, clients to serve, and operations to manage. Content creation was supposed to be a growth lever. Instead, it feels like another job.

One you’re losing at.

I know this because I lived it.

For months, I watched my Substack growth plateau. Not because my content got worse, but, ironically enough, because it was working well.

My calendar got fuller. More people in my orbit meant more conversations. More conversations meant more clients. More clients meant more work.

The exact success I wanted created the problem I couldn’t solve.

I had less time for the one thing that brought those people in: showing up consistently online. My posting frequency dropped. My visibility shrank. New followers slowed to a trickle while peers and competitors alike kept publishing.

I wasn’t losing because my ideas were weak. I was losing because I couldn’t scale them.

You write one deep article. It takes hours of research, thinking, and crafting. You publish it. Maybe it performs well. Maybe it doesn’t. Either way, by the time you catch your breath, it’s already old news. The value you created sits there, dormant, reaching only a fraction of the people it could help.

Meanwhile, you need to be everywhere: LinkedIn, Twitter, Substack, Medium. Each platform has its own rhythm, its own expectations, its own format. Creating native content for all of them? That’s not a full-time job. That’s three full-time jobs.

So you pick one. Maybe two. And you accept that you’re leaving reach on the table.

It doesn’t have to be that way anymore

I’ve been writing about building assets for months. Systems that multiply your effort. Tools that create value while you sleep. I even wrote about what assets you need to win in 2025, and buried in there was this line:

“I’m building a fully automated system for this right now using workflow automation. Not available yet, but when it is, it’ll turn one piece of content into many, while preserving your core concepts and unique thought leadership angles.” -- Why 90% Of Creators Will Burn Out In 2026

This is that system.

I’ve been tinkering with it for weeks, and finally finished it over the weekend. Now I can take one article and turn it into multiple pieces of content that live across platforms, reach different audiences, and extend the life of my ideas—without adding more work to my calendar.

No more choosing between depth and reach.

How I quietly solved it with AI

What I did was create a workflow with n8n, a low-code, developer-friendly platform.

This was an intimidating leap for me because I have zero coding background, but I’ve been wanting to for a while. And with 2026 just a stone’s throw away, I thought, no better time than today.

So, here’s what I did:

1. I identified my actual problem

So, I wasn’t publishing as much because I was busy. My first thought was that it’s a content creation problem. Creating new content felt urgent -- plus, I always feel like that’s a strength I have: creating new ideas and telling new stories.

But on the other hand, I didn’t want to create more new content … faster. I have a writing system I built for working on new ideas, but … I don’t want to automate that. That’s the deep work I love; organizing thoughts, theorizing, crafting narratives that stick. That’s where I stay unpromptable.

On the other hand, I have hundreds of articles sitting in my archive. Deep, useful, evergreen content that people still need. But they’re invisible. Dead links on old URLs, buried in newsletters nobody scrolls back to.

There was a wealth of information there. Plus, content repurposing is somewhat repetitive, formulaic. It takes time, but it doesn’t require the kind of creativity I want to preserve.

That’s where automation makes sense.

2. I tried other solutions first

Before building anything, I customized my Authentic AI system with a standard set of repurposing prompts. It worked—technically. But the process was clunky.

For every piece of content, I had to:

1. Input the full article into the LLM.

2. Ask AI to analyze it with specific prompts.

3. Prompt again for each piece of repurposed content.

4. Copy, paste, then manually tweak.

5. Finally, format then publish.

Creating four short-form posts from one article? Four hours. Turning that article into a LinkedIn carousel? Another hour. A thread for Twitter? More time.

The AI made the writing faster. But the process was still manual. Context switching drained me.

You can imagine why I needed something that handled the middle part—the analysis, the drafting, the formatting—so all I had to do was input, review, and publish.

3. I outlined my ideal solution

The automation needed to do one thing well: take my long-form content and break it into platform-specific pieces automatically.

Not just shorter versions of the same thing. Real adaptations. Content that felt native to each platform, aligned with how people consume information there.

Here’s what I wanted it to create from one article:

• Two Medium articles (300-500 words each, different angles)

• One LinkedIn carousel (caption + slides)

• Four LinkedIn posts (varied formats and hooks)

• Four Substack notes (digestible insights)

All I’d have to do: drop a link into a spreadsheet, let the workflow run, then review and publish.

4. I built it with n8n

n8n is a no-code automation platform. Think Zapier, but more flexible and developer-friendly. I’d been using it for client projects, so I knew it could handle complex workflows. Here’s exactly how everything is laid out:

A. The trigger

A Google Sheet. Anytime I add a new row with an article link, the workflow activates.

B. Content parser

Two nodes pull the content from the URL, clean it up, and make it usable. No messy formatting. No broken links. Just clean text ready for analysis.

C. Content branches

The workflow splits into four paths. Each one handles a different type of content: Medium articles, LinkedIn carousels, LinkedIn posts, Substack notes.

For each branch, two things happen.

First, the workflow loads examples. These are high-performing pieces I’ve published before—content that worked, content I want to replicate. For the Medium branch, it loads four examples of shorter articles derived from longer ones. Same for the other branches.

Second, it loads a detailed prompt. Too long to paste here, but the core instruction is this: read the examples, read the main content, analyze it, break it into specific points, then create new content mapped to those points. Make it platform-specific. Filter it through unique angles calibrated to what works there.

D. Claude Sonnet 4.5 node

This is the brain. I access it through OpenRouter, a service node that lets you tap into multiple AI platforms with one subscription.

I use Claude because it’s the best writing tool I’ve found. It handles huge context windows. It understands nuance. It doesn’t flatten my voice into corporate speak.

E. Content organization

Once generated, the content gets cleaned, labeled, and sent back to the Google Sheet. Each piece is formatted so I can immediately copy it into the platform. No reformatting. No rewriting. Just review and publish.

A true content flywheel

I finished building this over the weekend. It’s not perfect yet—I’ll need to tweak prompts, adjust platform-specific formatting, test different angles. But it works:

One article now becomes eleven pieces of content. Same ideas, different formats, different platforms. The deep thinking I do once gets distributed across weeks of publishing.

That’s not a treadmill anymore. That’s a flywheel.

My effort compounds. My ideas reach further. And I get my time back to do what actually matters: create new things, talk to people, build relationships.

I’m just scratching the surface of what n8n can do. In the next weeks, I’m planning to build more with n8n. Systems for my own growth, and hopefully, for yours too. Stay tuned.

---

P.S. Would you be interested in getting this n8n workflow for free (set up guide + JSON file)? I’m planning to turn this into a free lead magnet for other creators to use! Let me know in the comments.