Content Hub OS: Lab Notes — 03.27.26
In this Note: How the RSS to podcast pipeline works inside CHOS. Make pulls the feed, generates a script, ElevenLabs records it, and the episode is ready to edit.
The Automator’s Log documents Content Hub OS as I use it — what’s inside, how it works, and what I’ve learned running it on my own brand before opening it to anyone else. If you’re new here, CHOS is an AI-powered content operating system built for operators who are done doing everything manually. These notes are the inside look.
Started building the RSS-to-podcast pipeline about a month ago.
This one’s been on the list for two years.
I’ve always wanted to cover AI news with my own angle on it. The problem was never the ideas. It was the production time. Recording a three-minute episode still meant scripting, recording, editing, file management. More hours than a quick take deserves.
So I waited until I could build it right.
It didn’t go fully automated on day one. First few weeks were half automated, half manual while I figured out where the gaps were. Then figured out how to automate another 30% of it. That’s usually how this goes — you build what you can see, run it, and the next bottleneck reveals itself.
Here’s what’s running now: RSS feeds pull the news. Make triggers on new entries, summarizes the story, generates an outline and script in my voice, sends it to ElevenLabs to record using my voice clone, and drops the finished file back into Content Hub OS with the metadata attached.
About 10 minutes of my time to review and approve.
First episode went live March 3. 962 downloads in 24 days. Substack only — YouTube and Instagram aren’t in that number.
What took the longest: the prompt structure for the script generation. Generic in, generic out. Had to get the episode format locked and specific before the output sounded like me instead of a news recap.
Running it on my own brand for a few more weeks before it opens to users. I need to know my results can be duplicated before I hand it to anyone else.
That’s the only QA that actually matters.
✌️ Audra
Stop fighting your tools and start using a system designed for human behavior.”
Most workflows fail because they ignore the operator. I’ve built the Content Hub OS using the behavioral science and NLP frameworks I’ve mastered over 25 years. It’s designed to be frictionless.
See what a “hardened” system looks like in action.
Join the Founding Members - Here