I’ve seen a lot of automation workflows, but this one is next level. It completely solves the tedious task of content repurposing, and I think you’re going to want to build this for yourself! This incredible AI professional shared a full walkthrough of a system he built using Make AI Agent. The mind behind it created a workflow that watches a Google Drive folder, grabs new scripts, repurposes them for LinkedIn and YouTube Shorts, posts to LinkedIn automatically, and then Slacks him a link to the finished work.
This isn’t just a simple trigger-action zap. The real magic is how the creator uses the AI Agent as an intelligent conductor for a whole orchestra of apps. The agent receives a goal, analyzes the situation, and then decides which tool to use from a toolbox the creator provides. It’s a fantastic real-world example of agents moving beyond chatbots and into practical, autonomous work.
Here’s a breakdown of how this industry pro built this incredible content machine.
🛠️ How The Expert Built It: A Step-by-Step Guide
The creator’s process is super clear and logical. He essentially gives an AI agent a mission, the tools to complete it, and a trigger to start.
- Set Up the Agent & AI Brain: First, the expert created a new AI Agent inside Make. He connected it to OpenAI (you can use others like Anthropic or Gemini) by providing an API key. He recommends using a model like GPT-4o mini for a good balance of speed, cost, and intelligence.
- Define the Mission (System Prompt): This is the agent’s master instruction set. The author wrote a detailed system prompt telling the agent its role is a “Script Repurposing Agent.” He clearly defined the inputs (a Google Doc script) and the desired outputs (a YouTube Short script and a LinkedIn post).
- Teach It Your Style (Context): This step is brilliant. To ensure the AI writes in his voice, the creator first used ChatGPT to analyze one of his old video transcripts. ChatGPT produced a document detailing his tone, style, and reading level. He then uploaded this “style guide” as context for the Make agent, telling it to use that style for all repurposing tasks.
- Give It Tools (Modules & Scenarios): This is where the agent gets its power. The creator gave the agent access to several tools:
- A Multi-Step Scenario: He built a workflow to “Get Script Text from Google Docs.” This tool takes a File ID, finds the document, and extracts all the text for the agent to use.
- Single-Step Modules: He then added simpler, one-action tools like “Create a new Google Doc,” “Create a company post on LinkedIn,” and “Send a direct message on Slack.”
- Set the Trigger: Finally, he built a separate automation to kick everything off. This scenario’s job is to watch a specific Google Drive folder. When a new file is added, it triggers the AI agent, hands it the File ID, and says, “Repurpose this content.” The agent then takes over, using the prompt, context, and tools to complete its entire mission.
🧑💻 Who is this for?
This type of automation is a perfect fit for anyone who creates pillar content and needs to distribute it across multiple channels. I think it’s especially powerful for:
- Content Creators: Turn one video script or blog post into a dozen social media updates without lifting a finger.
- Social Media Managers: Automate the creation and posting of derivative content, freeing you up for strategy and engagement.
- Marketing Teams: Ensure consistent messaging and tone across all platforms by standardizing the repurposing process.
- Small Business Owners: Save countless hours by putting your content marketing on autopilot.
⚠️ Potential Pitfalls to Watch Out For
The creator’s setup is solid, but when building something similar, it’s wise to be aware of a few things:
- The Infinite Loop: The author smartly saves the agent’s output files to a different folder than the input folder. If you don’t, the agent will see its own creation as a “new file” and try to repurpose it, creating a costly and messy loop.
- Unexpected API Costs: Remember that using the OpenAI API isn’t free. While individual runs are cheap, costs can add up if your automation runs very frequently or processes extremely large documents. Keep an eye on your OpenAI billing dashboard.
- Brittle Prompts: The agent is only as good as its instructions. A poorly worded prompt or an unexpected change in the input format can cause the entire chain to fail or produce strange results. Test thoroughly and be prepared to tweak your prompts.
- Connection Problems: This workflow connects many services (Google, OpenAI, Slack, LinkedIn). If an access token expires or a permission is changed, the automation will break. The new Make Grid feature mentioned by the creator is a great visual tool for debugging these connection issues.
✍️ Prompt of the Day
The person who shared it was kind enough to include the exact system prompt he uses to command his agent. This is a great template for building your own.
You are the Script Repurposing Agent. Your goal is to repurpose YouTube scripts that I write in Google Docs into two new formats:
- A 60-second YouTube Short script with a strong hook, three quick steps, and one clear CTA.
- A LinkedIn post version written in the same voice, using 3–6 short lines or 1–2 short paragraphs.
Here are the steps:
First, you will be given the Google Doc link to the most recent YouTube script, which you will use in subsequent steps.
Then, get the full text of the script document using the “Get Script Text” scenario and the File ID attached to the Google Doc link given to you in the original message.
Then, repurpose that into a YouTube Shorts script and ready to post LinkedIn Post.
Then at the end, using the “Create Document for Shorts…” Module, put all of the repurposed content into only one newly created Google Doc.
Finally, send only the LinkedIn specific repurposed content off to LinkedIn and post it to our page.
When they are ready, send me the message on Slack with this new Google Doc.
🤖 Tools of the Day
This workflow is an excellent showcase of connecting best-in-class tools:
- Make AI Agent: The central brain and orchestrator of the entire process.
- OpenAI: Provides the core intelligence for rewriting and repurposing the content.
- Google Drive & Docs: Serves as the content repository: the source of original scripts and the destination for the repurposed versions.
- LinkedIn: The automated distribution channel for the final post.
- Slack: The notification system to confirm the job is done and provide links for review.
I was seriously impressed by the thoroughness of this setup. It’s a powerful, practical demonstration of how AI agents are ready to take on complex, multi-step tasks.
To see the full step-by-step video and grab the complete prompt details for yourself, you have to check out the original post from this talented creator.