Skip to content
How to Build Your Own Agent OS FREE artwork

How to Build Your Own Agent OS FREE

AI News Today | Julian Goldie Podcast

September 15, 2026

How to Build Your Own Agent OS FREE

Speakers Julian Goldie

TopicsMarketingBusiness

Julian Goldie (0:00)

So, agentic operating systems are some of the most powerful things you can build for your business and, you know, for personal stuff in general, simply because you can quickly have all of your agents in one place, you can operate them, you can build custom workflows. So, for example, we have like a video editor over here, we have a memory galaxy, we have, for example, an SEO agent. And these are all amazing things that you can build into this. But most people think, oh, you need to pay for the APIs to do all this sort of thing. Actually, today, I'm going to show you how to build an agentic operating system for free. And we'll just build it from scratch, and I'll show you some example prompts and stuff that you can build with this, and also show you how to operate it for free as well. So once you built it, you can actually plug in free systems into your agentic OS to keep using it for free. So let's get straight into this, and I'm going to show you the first place I would start. So what you can actually do is you can use a free open source harness to build this app, right? That's step number one. So for example, if you are interested in building out an agentic operating system for free, I would start with DeepSeek Harness, or I would go with something like Hermes Agent. Both of those are like free agentic harnesses.

You're free, they're open source, and you can build some cool stuff with it.

Now, you might say, okay, that's great, the harness is free, but what about the model? How do you get a free model for these systems? So you got a few options here. Number one is that you can use something like OpenCode as the brain of your agentic harness, and you can use this inside Hermes Agent. You can also use this inside, for example, Deepsea harness. All right. So if we break this down into steps, step number one is you're going to build the agent OS. I'm actually going to write this down, I think it'd be better written. How? You can use Deepsea harness or Hermes Agent. Both of those are free agentic harnesses. And then you're going to ask, well, how do you get a free model to plug in to your agentic harness? You can use open code or you could use something like a free API. So for example, open router, if you type in free here, there's loads of different free models that you can plug into your harness. Right. So for example, you've got Lightning 3 Flash, you've got, for example, LFM 2.5, you got all of these different free APIs you can plug in, which is pretty cool. So you've picked your free harnessing model, that is stage number one.

Now stage number two is you want to start building. So where do you start? Well, let's say, for example, you're using open code with free APIs to build out your agentic OS. You would go inside a new session here, right? So you're going to start a new session, and then you're going to say inside here and you'd select whichever free model you want to use. So for example, you've got Ling 3 Flash, you've got MIMO v2.53, we have Muspah 1.33, and we have Nemetron 3 as well. If you just type in free inside open code here, you can see all the free models you can use. Nice. So if we use MIMO v2.53, we're just going to say, can you build out a mission control dashboard where I can see open code running, and I can also build beautiful automations inside there. Should just be like a web app that's hosted locally, where we can build out our AI agents and have like a mission control for all our AI agents, and later I'll build out some custom automations inside there. Make it look beautiful, clean, interesting. So you basically do this, as you can see, and that is the example prompt that you can use. All right. So you're going to just go into your agentic harness, could be open code, could be deep-seed harness, could be Hermes agent. They all do the same thing in different ways. Then you can go into your agent and you can see here that it's beginning to build out step by step.

Now, once that's done, it's going to have a local host for your agent OS. What do you do from there?

That brings us on to step number three. So step number three is you want to build free models into your agent OS and customize it. Let me show you what that looks like. So first of all, your agent OS is going to be super basic at first. It's not going to look as good as this. It's probably just going to have the mission control dashboard in, and then also open code. So what you want to do is just go into anything that you've built inside your agent OS. So if we go to, for example, DeepSeek Harness over here, we can then give it some prompts and just test out if it actually works.

4 more minutes of transcript below

Thousands of transcripts fetched by people building searchable podcast archives

Fetch the whole transcript

The demo key returns a sample episode in full, no card needed:

request
curl -H "x-api-key: pt_demo" \
  https://spoken.md/transcripts/1000651996090

Markdown with the speakers named, for your notes, your knowledge base, or anything that makes HTTP calls.

From $0.10 per transcript. No subscription. Credits never expire. Prices exclude VAT, added at checkout for EU customers. Not what you expected? Email us within 14 days with 20 or fewer credits used and we refund the pack in full.

Using your own key:

request
curl -H "x-api-key: YOUR_KEY" \
  https://spoken.md/transcripts/1000789781231