Content Hub OS: Lab Notes - 04.09.26
In this Note: How a podcast transcript enters CHOS and generates descriptions, pull quotes, key points, title ideas, and a full supporting asset stack.
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.
The Podcast Workflow
Most podcasters promote their episodes once.
Maybe a few posts. An email to the list. Same angle, same clip, same caption across every platform. Then they move on to the next episode and everything from that recording disappears into a folder somewhere.
Not because they don’t care. Because pulling more out of it manually isn’t worth the time it takes.
This is where that changes.
Start by adding a new podcast episode. The form asks for the basics. Working title, recording date, brand, intended audience, content goal, content style, show name, episode title, and host name. If you have a guest select them from your saved guest list. If their profile is already in the system their bio and background load automatically. The AI uses that context when generating content so you’re not starting from scratch explaining who they are.
The most important field is the transcript. Paste it directly for episodes under 30 minutes. Upload a file for anything longer. That transcript is the source everything else gets built from.
Then select your foundation prompts. There are five:
A description for your show notes. A guest pull quote ready to use. Mentioned resources, tools, books, and links referenced in the episode. Key points pulled and organized from the transcript. And title ideas if you need options.
You don’t have to run all five. Pick what you need and come back for the rest later.
Choosing those five was the hardest part of building this workflow. Not technically. Curatorially. Running everything at once caused the automation to time out. Too much information moving through the system simultaneously. So the decision became which five outputs matter most as the foundation.
These five are the answer to that question.
Once the foundation prompts are done you move into the supporting asset stack. Social posts, emails, newsletter, images. Everything that promotes and amplifies the episode across platforms. Different formats, different angles, built from the same source so nothing is starting from scratch.
All of it lives in one connected record.
That’s the part worth stopping on. Three months from now you find a social post you want to rerun. Everything connected to it is still right there. The episode, the description, the pull quotes, the emails, the images. You don’t need to remember what blog went with what video or which email promoted which episode. It’s all connected to the source it came from permanently.
Most content operations are a graveyard of disconnected files spread across four platforms and a Google Drive folder nobody can navigate. CHOS keeps everything tied to the pillar piece it came from. That’s not a feature. That’s how content should have always worked.
✌️ Audra
One recording. Every asset. All connected. That's not a content strategy. That's a content system.