Speaker A
We all know Fable 5 was one heck of a model. Then it got pulled.
Learn how to build agentic harnesses to achieve Fable 5 level AI results with any model, ensuring autonomy despite model shutdowns.
Key Takeaways
- The power of AI depends as much on the harness as on the model itself.
- Building your own harness ensures autonomy and resilience against AI service shutdowns.
- Agentic harnesses enable long-term, autonomous, creative AI workflows.
- Frontier models are best for building harnesses; open source models are ideal for economical execution.
- Owning the harness gives you leverage and control over your AI workflows.
Summary
- Fable 5 was a powerful AI model that got pulled, but its success was due to the agentic harness (Cloud Code) it used.
- An agentic harness acts as the body and tools for an AI model, enabling long horizon autonomous creative work.
- Harnesses can be built to work with any model, including GPT, Gemini, Opus, and open source models.
- Owning the harness (rig) rather than renting the model intelligence protects against disruptions like government shutdowns.
- The AI agent loop involves reading prompts, selecting tools, running them, verifying results, and iterating until done.
- Building your own harness allows you to plug in frontier models for quality and open source models for cost efficiency.
- The harness includes context (knowledge base), tools, verification systems, memory, and decision logic.
- Using frontier models to build harnesses and open source models to run them balances cost, privacy, and performance.
- The video demonstrates building a harness for a directory project, showing practical steps and workflows.
- The creator encourages viewers to build sovereign AI systems to avoid dependence on any single AI provider.
Chapters
- 00:00Introduction to Fable 5 and agentic harnesses
- 01:22Fable 5's directory project and its capabilities
- 02:45Components of an agentic harness
- 03:57Owning the harness vs renting AI intelligence
- 05:07Importance of harnesses for AI resilience
- 06:16Using frontier models to build and open source to run
- 07:41Demonstration of harness capabilities and workflows
- 10:32Building your own harness step-by-step
- 12:22Harness management, content production, and project integration
- 14:40Final thoughts on sovereign AI and harness ownership
Full Transcript — Download SRT & Markdown
Speaker A
But if you know how to build a system that makes any LLM perform at that level, then you'll never have to worry about chasing the next big AI breakthrough or accompany your government rug pulling you ever again.
Speaker A
And that's what I'm gonna share with you in this video: what an agentic harness is, how they work, and how to get started building your own so you can get long horizon creative autonomy without Fable 5.
Speaker A
Plug in any model you want. So in this video, I'm gonna introduce you to what an agentic harness is.
Speaker A
I'm gonna show you one of mine in action right now, and then we're gonna build one on our own using a template or simply prompting our AI to do it.
Speaker A
And then I'll give you some more tips on getting started. Before we begin, I just wanna encourage you to hit that subscribe button because this type of sovereign AI usage, like the real foundations of AI, is what I like to
Speaker A
explore on this channel. I don't think anyone should be dependent on the shiny objects, the next big hyped tool out there, the next big hyped model.
Speaker A
We should know how to use this intelligence from the ground up. So when the rug gets inevitably pulled out from under us and we lose the best model ever, we're not completely screwed.
Speaker A
Just a few hours before Fable was pulled, I built a 240 listing directory. Fable came up with the SEO, AEO strategy.
Speaker A
It scraped the whole web, and it created the entire architecture and infrastructure of this sovereign individual-minded directory.
Speaker A
It categorized things. It made an llms.txt. It rendered the whole thing with Astro, and it was just a really impressive result, so I got really excited about Fable.
Speaker A
And then Fable was pulled, and builders, including myself, were really bummed because it did such a good job at long horizon autonomous work.
Speaker A
But the truth is, we don't actually need Fable to do long horizon autonomous work.
Speaker A
Opus, other models are all intelligent enough. What you need is a good harness. An agentic harness is the rig that keeps the building loop for an AI going.
Speaker A
When you know how to build your own harnesses, you're not gonna be as vulnerable to the rug pulls that might happen due to a government or a company or whatever it might be next.
Speaker A
So in this video, I'm gonna show you all about agents and harnesses, how they work, and then we're gonna set one up to keep building my directory for the next steps because I can't use Fable anymore.
Speaker A
So let's talk about the loop. Now, what an AI is gonna do when it's building an agent, it's gonna read what the prompt is.
Speaker A
It's gonna pick a tool, it's gonna run the tool, it's gonna check it, and then it's gonna find out if it's done or not, and then it's either gonna start over or move on to the next thing.
Speaker A
That is what the AI agent loop looks like. Okay, so let's unpack that. You have AI, the intelligence, the LLM, that is the brain.
Speaker A
Okay, that's here in the middle. But the harness itself is going to be the body, the fingers, the arms.
Speaker A
What it needs to actually get stuff done, that actually is good. So you have the context, which is like a knowledge base.
Speaker A
You have the tools, which might be MCPs or using the terminal. The harness has systems for verifying that the work is good.
Speaker A
It has systems for determining if the work is done or not. And then it has a memory, like what actually happened and what needs to be next.
Speaker A
So in essence, an AI agent is really a model plus a harness. And then this is really important to remember, in my opinion, because Fable was the model, okay?
Speaker A
The reason Fable was so frickin' good, because the Cloud Code or Cloud App is the harness that makes the agent so frickin' good.
Speaker A
So when you don't have Fable anymore, what do we do? Don't fret, because we can create this on our own and in some cases even make it better.
Speaker A
So a good harness can run with any model. You can run it with GPT, you can run it with Gemini, you can run it with Cloud.
Speaker A
And perhaps most importantly, you can run it with open source models. The model is probably not gonna be run on your local machine, though it might be able to.
Speaker A
Especially maybe a year from now, it might be completely normal that we run harnesses on our machine.
Speaker A
I've been using Goos a bit with local models and Odysseus, and we're getting to a point where this is gonna be a thing.
Speaker A
But for now, what's important to understand is that while we're renting the intelligence probably we can own the harness.
Speaker A
And I'll also call it a rig in this video, because that's what we call in the AI Captain's Academy.
Speaker A
You can own the rig, which is like the software, the body, the robot, that you can plug the intelligence into.
Speaker A
And if you can own that, then you're not as fucked when the government tells Anthropic to shut down its model.
Speaker A
Okay, so let's address the two problems with letting the model itself hide inside the harness, which isn't exactly what Fable was, but for sake of the metaphor here, let's go with it.
Speaker A
Problem number one is you rented it. So perhaps it's not Fable 5's intelligence that made it so impressive.
Speaker A
Obviously, Opus 4.8 could have made that directory, but the harness, Claude Code, which Fable 5 was designed to really use and maximize, was the body for Fable.
Speaker A
So Opus can still use Claude Code. It's just not gonna be as good with it unless you set it up to be as good as it can be, which we're gonna do later.
Speaker A
So imagine the harness lived inside of Fable, you rented all that. So if it's no longer available, it's gone.
Speaker A
Problem number two is you have no leverage when it's not yours. And this is the main point I'm trying to drive home here, is you wanna know how to build these things to plug in any model, whether it's local, remote, whatever.
Speaker A
So then the next time a big black swan event happens, you're not screwed over.
Speaker A
I'm just gonna keep saying that. The agentic harness is your ticket to running powerful AIs without worrying about where the intelligence is coming from.
Speaker A
In a previous video, I mentioned Odysseus. That is a great workspace. It's not quite a harness in the same sense we're talking about here.
Speaker A
Here we're talking about being able to one-shot entire projects, long horizon, autonomous creative work.
Speaker A
So you can build the harness yourself and then run it with pretty much any model.
Speaker A
Now, an agent, like I mentioned, is essentially a model plus a harness. You can still plug a weaker brain into a great harness, and it'll still ship.
Speaker A
And it'll still do a pretty good job the better your harness is. You could have a genius brain, but no harness, and it's still just gonna be a chatbot.
Speaker A
It's not gonna do anything. It needs to have the agentic capabilities to do stuff.
Speaker A
So to me, this is like the elephant in the room. Fable 5 was amazing because of how it works with Cloud Code.
Speaker A
So what I've been saying for a long time in the AI Captain's Academy is you wanna build your tools, your workflows, your pipelines, your harnesses, with frontier models, and then execute with open source.
Speaker A
Why? Because with the frontier models Opus GPT Gemini it's gonna cost more, but it's gonna do a great job.
Speaker A
Even in Odysseus, you have Teacher Mode, which will call a frontier model to create a skill or procedure that the open source model just doesn't do well.
Speaker A
But then, once you've got a quality harness built by the frontier models, not only is it yours, but now you can run cheaper, more economical open source models, either local or remote, and get better privacy and save money,
Speaker A
and get really good results because the harness is really good. I use the Venice API for my harnesses because I can call frontier models and I can call open source models and I can choose for which task am I gonna need
Speaker A
which model for either saving money or making sure I get the best performance possible for that specific task.
Speaker A
So what this means in essence is AI writes the code, now it makes the things for you, will you just say what you want?
Speaker A
Your job has moved up. You are now the strategizer, you are the director. The AI does the labor.
Speaker A
The
Speaker A
You want good quality, you wanna make sure you don't have to babysit it. Once you get these harnesses created, you plug in whatever the input is, in my case, like a transcript.
Speaker A
So let's dive into what it means to build a rig from scratch, okay? You don't really need to start one from scratch.
Speaker A
You could, Cloud Code and Codex know how to do it, but I actually recommend stealing a good one.
Speaker A
So this is from CheddarFox, Scott Graham, who was on the channel a while back to introduce these and changed my whole life learning about these.
Speaker A
So this is the Safe Agentic Workflow. This is for development and we're not gonna need something for all cases that is this crazy, but you could use a harness like this to build software.
Speaker A
This is the Scaled Agile Framework Methodology, Adaptive for AI Agent Teams. You could use it for any kind of team.
Speaker A
So let's explore what that actually means. So what happens in the harness is you have multiple agents in a team, in this case you have 11 agents.
Speaker A
They all do different things, they all have skills, they all are triggered based on hooks or the main orchestrator agent calling them.
Speaker A
Now I am not a full-on developer or whatever, so I'm not gonna get too deep into this because I'm not the person to be able to do that.
Speaker A
But just imagine you have all these skilled agents on your team that know what they're doing and have a full fresh context window to do it.
Speaker A
Here's an example of a harness I'm running right now inside Cloud Code. This is inside my brand knowledge base and it's already been running for seven and a half minutes, maybe not too long, but it's got a bunch of phases here
Speaker A
where it takes a transcript from a video or conversation and turns it into all kinds of content and makes images, it makes a sub-stack draft, it makes a ConvertKit draft, it reviews it all.
Speaker A
It does everything, it calls different models to do that, like it uses Opus to write but then uses GPT 5.5 to review what was written and make sure it fits all the brand guidelines and then it makes videos,
Speaker A
which you're gonna see in this video. Anyway, this is a harness here and this just goes on for as long as it needs to and it creates whatever I set it up to create.
Speaker A
You don't have to be a developer or a fancy engineer person to be able to use something like this.
Speaker A
At the end of this video, we'll jump into an actual template on GitHub that you can use to get started and I'll show you some specific templates that help with certain tasks and how to build your own.
Speaker A
Okay, so let's go back here, I accidentally closed it. Whoops, it was at about 21, 22 minutes, so it's gonna keep going now, it's gonna start over.
Speaker A
But anyway, if I hadn't closed it by accident, we'd be at over 22 minutes now without needing any input by me.
Speaker A
So actually, this is actually a good example here. SendMessage isn't available as a tool here, I'll spawn a fresh agent for the cycle.
Speaker A
For the next part of wherever it's at, it needs to use the SendMessage tool, but the main agent can't do it, it has to spawn a new agent that does get access to that tool.
Speaker A
If we wanna see everything, it spawned this sub-agent that says, okay, you are the repurposed threads agent, revision cycle two on the standalone tweets only.
Speaker A
So read this file and then reread the skills, thread writer, voice standard and stop-slop and make sure that it's good, basically.
Speaker A
So it's got its constraints here, it has its job. Okay, so let's back up.
Speaker A
What just happened here is in my workspace, the Jordan-Herbs brand workspace, we have my knowledge base, which says everything about basically who you are, who I'm making videos and content for, how I talk about things, my glossary sovereign-preneur
Speaker A
the sovereignty ladder, which you saw in the directory, my brand positioning, what is my brand, who do I speak to, what am I writing about, the thesis, my voice in tone, anyway.
Speaker A
So all of this stuff, my frameworks, everything about my brand is here, as well as my sub-projects, as well as my domain.
Speaker A
So whenever I open up an agent in this folder, I can literally just say, hey, update my website with this and it has everything it needs and it knows what to do.
Speaker A
So that's part number one, that's the context, my knowledge base. That is a crucial part of a solid harness, an AI agent that needs to have its knowledge base.
Speaker A
Then comes the .cloud folder. Inside the .cloud folder is where we're gonna see, okay, these agents being spawned.
Speaker A
In here we see we have our main agents, blog writer, content ideater, and then this is like their system prompt.
Speaker A
Content ideater, you generate content ideas for the sovereign-preneur brand across all formats and map them to the four-tier sovereignty stack.
Speaker A
First, read the thesis, then read the four tiers, then read the positioning document. Okay, so it says here's all the things you need to do, catch up on the knowledge, here's what you gotta do.
Speaker A
But in this case, we're in the content folder, this one, and so we have another .cloud folder, which says agents repurpose threads.
Speaker A
So this agent produces X content from a transcript analysis. And here we are in there.
Speaker A
So in my main folder, it's got my agent system prompts, and then in my actual content repurposing folder, it's got its agent system prompts.
Speaker A
Then you have the skills. So we have the newsletter writer skill that knows how to write like me and says, "Hey, check out the voice and tone document "to write like Jordan." It's got the stop-slop skill, props to Hardik Pangea on
Speaker A
GitHub for creating this, like, "Hey, make sure it doesn't look like AI," basically. And it's just got the skills that are needed for that workspace.
Speaker A
And then here we have no skills at all, actually, because it's actually pulling the skills from here, but we do have commands.
Speaker A
So the repurpose command repurpose a transcript into a full Jordanerbs brand content package. So this is basically what the whole orchestrator agent is following in this harness workflow.
Speaker A
And that's what keeps it on track here. But this is the beginnings of a harness.
Speaker A
You have the agents, you have the skills inside a workspace. And then in the safe agentic workflow, you'll also have hooks, commands agents skills and even a Claude MD.
Speaker A
So we didn't talk about Claude MD. It's also called agents MD and cursor. But essentially the Claude MD here gives the context for the conversation to your main agent.
Speaker A
So in our case, Claude MD, this is the Jordanerbs brand workspace. So when I start Claude here, it automatically says, "Hey, this is the Jordanerbs brand workspace.
Speaker A
"Open this folder when working on anything "related to the brand." And so it knows what all the folders are.
Speaker A
And it just has to read this file, Visual Generation Workflow, Plain Language Standard, Voice Rules.
Speaker A
It starts here, so it doesn't actually have to load every single file. It already knows here in the beginning, hey, when you're looking for brand specific skills, go here.
Speaker A
When you're looking for the content repurposing harness, open up this. And that is how this main agent, when Claude code is opened on the command line here and VS Code, will automatically know, okay, Jordan provided a transcript, so let's run the multi format content
Speaker A
repurposing harness. And that's exactly what's going on here. And like I mentioned earlier, we're at 21 minutes.
Speaker A
We're now approaching 30 minutes here with no interaction from me, aside from when I closed it by accident.
Speaker A
And it is just running its thing. And it's got code it runs. It's got skills it touches.
Speaker A
It does all kinds of stuff. Like imagegen.shell, this is a script here. Let's look at the prompt.
Speaker A
Yeah, this is the prompt with my colors based on the brand, which is here in Jordan Herbs KB.
Speaker A
We see it here in design, design system. There's the Jordan Herbs visual design system.
Speaker A
And so this whole harness knows where to look for all the data without having to overwhelm a single agent.
Speaker A
Because if one agent were to load all these documents, it would take hundreds of thousands of tokens.
Speaker A
And you only have a million token context limit and your quality gets worse the more tokens that are used.
Speaker A
So instead, the main agent only loads what it needs to, if it needs to.
Speaker A
And otherwise, we'll spawn an agent, give it the agent system prompt. Like, hey, you're the content ideator, start doing this.
Speaker A
Here's what you do. And here are the documents you're gonna need. So in this case, the content ideator is gonna need to read the thesis file, the sovereignty stack file, the positioning file, the glossary file, the ideation protocols,
Speaker A
which is all quite a bit. At least it takes place in a sub-agent conversation, a fresh new context window.
Speaker A
So it doesn't have to clog up the main one, because the main one is just making sure we get everything high quality.
Speaker A
So then after that, whatever agent, let's say in this case, we have the image generator agent, whatever it's called in here, generating the assets, then another agent's gonna come through.
Speaker A
Well, let's say, for example, the thread was written, then the voice reviewer agent is gonna come through.
Speaker A
In a new conversation, doesn't know anything about the past, and the main agent says, hey, here's your system prompt, you're the voice reviewer.
Speaker A
First steps, check out the voice standard skill, the stop slop skill, the messaging, document, voice and tone.
Speaker A
So learn everything about Jordan's brand, how you write, how to not write, and then you're gonna check what was written by the last sub-agent in a conversation you have no idea about, you have no context of.
Speaker A
You're gonna check that out, and you're gonna make sure it's good. And then here is the return format.
Speaker A
You're gonna give a report. Then that report goes back to the main agent, and the main agent determines, okay, do we go on or not?
Speaker A
And that brings us back to the loop. Read, pick tool, run, check, are we done?
Speaker A
Yes or no. And then it just keeps going. So essentially, every sub-agent is running in this loop in some way, shape, or form.
Speaker A
We're running everything right now to Cloud Code in Opus. But the great part is once this is built, and the same thing with this here, I can open this folder in Open Code using an open source model here via Venice,
Speaker A
and I can run the same workflow, and it will do pretty much the same job, just with a different intelligence model behind it.
Speaker A
So if Cloud Code ever rugs me, and I can't use Opus anymore or whatever, I have a rig that I can just plug in my custom intelligence to.
Speaker A
So the model, yes, is rented, but the rig itself, the harness, is owned. This is the sovereignty filter.
Speaker A
I like to call it, it's in the framework in my knowledge base. You can swap the brain, but keep the body.
Speaker A
You own the body. You are the sovereign preneuer here. So now is the time I'm gonna give you a CTA in the AI captain's academy.
Speaker A
We have a bunch of agentic harnesses for solo preneuers to kickstart your next venture, especially if you're trying to sell AI.
Speaker A
So for example, this is a lead getter. It'll use Apify to scrape the web and find leads for you.
Speaker A
It'll verify the contact info. It'll find the specific pain signals, and it'll even create a hook for your cold email.
Speaker A
So it's a funnel, there's a little video explaining how it works here, and it walks you through step by step.
Speaker A
So in this case, it'll help you with a client. You could also use it for yourself.
Speaker A
The idea is that you run an audit on a website, and it will run this 10-agent pipeline, web crawler, API caller, to call free APIs for metrics about the website.
Speaker A
It'll find competitors, it'll analyze AEO, it'll strategize for SEO, write a report, design the report, build the report, review the whole design, then write a cover letter, and then validate everything with quality review.
Speaker A
And then you can just email the client and be like, "Okay, hey, I made you this AEO audit.
Speaker A
"This is free, but if you want the big one, "it's a few thousand bucks, but here it is." So we have all these rigs in here, they're all harnesses.
Speaker A
And they're free, it's seven-day free trial, and come in here and steal them, and then you can leave, it's all good.
Speaker A
This is my personal content forge. This is exactly what I was showing you here.
Speaker A
This is it. Oh, hey, so yes. It created a sub-stack draft. Boom, premature freedom.
Speaker A
I built the dream, then gave it up on purpose, right? And then it wrote a newsletter draft.
Speaker A
It was sent to the wrong place. That's interesting. So it made everything, it baked for 16 more minutes.
Speaker A
So all of that took about 35, 37 minutes to run through everything. Here's all the content, premature freedom.
Speaker A
It's funny. It's kind of a random one, but we got the YouTube metadata here, titles, description, even though this is not a YouTube video yet, chapters.
Speaker A
I normally run this with the YouTube transcript after this time I ran it with the transcript from before.
Speaker A
Then here we have an X thread with images that weren't generated. Oh, there were images.
Speaker A
Okay. They're generated images on brand. And then we got the blog posts here. We got the emails here, the newsletter.
Speaker A
So it went through and did all that basically. That's a harness. And now I'm just going to say, fix that part of the harness so it doesn't happen again.
Speaker A
And then get push. The harness is fixed. It sent the newsletter to convert kit right here.
Speaker A
I built the dream then moved to a suburb on purpose. Yeah, it is kind of weird what I did with my life.
Speaker A
And then finally, probably the coolest bit is here on the web now, premature freedom.
Speaker A
Oh, it didn't pass the test. This is for my ghost writer, which is another video on the channel.
Speaker A
Hopefully you'll see, but this is the UI for my ghost writer because I like making the content, but I don't like posting it.
Speaker A
So this is the X thread that can be copied or you can edit it.
Speaker A
I don't make the blog posts. So it's got an image here all on brand, the newsletter, which is what we just saw on kit and then all YouTube metadata, which isn't accurate at all, because this isn't actually a YouTube video.
Speaker A
Like it normally is. And then there's the checklist of everything that's here. And then the ghost writer, Tommy can mark things as posted.
Speaker A
It all syncs to get hub and boom, that's a Jordaners content repurposing system. That's a harness in action for something.
Speaker A
I think every solo, pre-new or creator kind of should probably be valuing repurposing your content without needing a hundred different subscription platforms or a big team, et cetera.
Speaker A
So how do we get started? Well, the first step would be simply to copy something that's already been made.
Speaker A
Now you have the safe agentic workflow and you could simply use this template. You could copy the URL.
Speaker A
There's a link below, by the way, copy it. And then in your IDE or even open code, which I've been becoming a fan of the desktop app lately, you start a new session.
Speaker A
You say, let's build a harness based on that URL. I want it to do X and Y and Z.
Speaker A
Let's talk about it. And then shift into plan mode and then you'd get started.
Speaker A
That's the like simplest way to do it. And your AI will then say, okay, we should duplicate it.
Speaker A
We should clone it. Or we should just take the ideas and build our own thing.
Speaker A
Every model is probably going to have a different opinion on that. So let's explore how I would do this now for the sovereignty directory.
Speaker A
So here in the sovereignty atlas.com, I noticed that my data for SEO API key wasn't being called like I thought it should be.
Speaker A
So I said, Hey, run an audit. What's going on with our SEO? Was the strategy made correctly or not?
Speaker A
And basically it said it was. But Opus came back saying, all right, here's what we can do to fix the schema and make it better.
Speaker A
So my recommendation for the next step is still a seed the listing facts, deploy the schema fixes now, or C build.
Speaker A
I want to do all these things, but I want to do them efficiently. I want us to build a harness for these common tasks of adding new content and ensuring that the schema and everything matches and fits the directory
Speaker A
strategy most optimally, and then I'll go into plan mode here. Let's talk about it.
Speaker A
Sub agents, skills, et cetera. So now I didn't copy any GitHub URL. I just said, let's build a harness.
Speaker A
That's a great instinct. Right. Thanks a lot. The work I just did by hand is exactly the type of repeatable workflow that should be a harness.
Speaker A
There you go. So we're now building a harness inside cloud code for this repetitive task for this particular project, which is a directory and it's going to need new content.
Speaker A
So this will save tokens and time in the long run, because next time that I want to add new FAQs or new pages, like the home setting libertarian pages, it'll just run through the harness and it won't need to ask me any questions and I won't
Speaker A
need to waste a bunch of tokens, like figuring it all out like it will right now.
Speaker A
Okay. So now Claude has come back to me, which content workflow should the harness cover?
Speaker A
This determines how many commands and agents we build. Seed frequent ask questions on listings.
Speaker A
Yes. Add a new listing. Yes. Write a new guide. Yes. Refresh research and audit.
Speaker A
Sure. Why not? How autonomous should the pipeline be? Where does a human gate it?
Speaker A
So basically I'm just going to need if with this recommended option, the agent re researches drafts and reviews the voice.
Speaker A
The orchestrator presents the final artifact. My approval before it touches seed files or the DB.
Speaker A
I'm just going to say fully autonomous because I'm pretty confident with Fables original work.
Speaker A
Where should the harness live and how should content reach production? Yeah, right there. Great.
Speaker A
So now it's building the harness and we'll come back to that and run it when it's done building because I didn't want to do the human approval.
Speaker A
It's going to do a automated quality bars that make autonomy safe, which is probably smart, but this is essentially like even this safe agentic workflow.
Speaker A
This is like enterprise grade AI stuff. These harnesses are how you go from like novice amateur AI slop to like a really good quality.
Speaker A
Now I don't post anything without reviewing it. Even the Atlas, I went through most of it before I put it online and I was like, holy moly, this is so good.
Speaker A
I was like so impressed with this research and everything. And I know most of it's accurate because I'm familiar with most of it anyway.
Speaker A
And I think that's just part of the appeal of Fable as well. And these newer frontier models.
Speaker A
But anyway, it does such a good job that then the human almost just needs to like read it.
Speaker A
Okay, that's good. Ship it. Uh, and you know, I still haven't gotten to that point with my blog posts and that's probably a good thing because they're supposed to be coming from me, a human, but they are a good starting point.
Speaker A
Okay. So it's pretty much done now. What you have now, a fully autonomous content harness at sovereign directory fast, Claude, so here we are at the Claude folder.
Speaker A
We have agents. Maybe it explains it here. We have commands for adding a frequent last question page schemas to listing up pages as a listing with the correct running and trust label writes a guide and Jordan's voice targeting a keyword
Speaker A
cluster or freshers demand data runs the SEO, AEO audit. Wow. Cool. Okay. Two open items for you to do populate the Netlify build hook URL.
Speaker A
Okay. I got to figure that out. Confirm the harness can reach the super based production connection.
Speaker A
All right. So want me to commit the harness or take it for a real spin?
Speaker A
We'll say yes. Let's do that. And to make sure it doesn't write, we'll say, write a handoff prompt to start in a fresh conversation.
Speaker A
So when you run a harness, you generally want to make sure it's in a new conversation, so there's no like leftover residual context and tokens.
Speaker A
And when you start fresh, you're going to get better results anyway. So here we get finishing up whatever it needed to do.
Speaker A
Going to make some commits and then we'll start a new conversation. So we got this handoff prompt.
Speaker A
We'll clear this and let's just read it real quick. You're in this directory. It has an autonomous content harness that I built in verified end to end.
Speaker A
Read cloud MD first. Actually, I don't even want to say all that state what I want to do in this session.
Speaker A
Run refresh research. All right. So I'm just going to copy these. Actually, we need these constraints too, because I haven't fixed whatever that is.
Speaker A
All right. So we'll clear this. Paste that. And now I'm not telling it that it's in a harness.
Speaker A
That's the real point here, because it should automatically open up cloud MD, which says, Hey, this is the sovereignty Atlas directory harness.
Speaker A
Here's everything you need to know about it. Here are your commands. Here are your agents.
Speaker A
Here are your skills reference files. So that's what it's going to do first. And you really want to make sure you have auto mode on too, by the way, but be careful with auto mode.
Speaker A
So it's running the refresh research skill. Let's check that out. Interesting. Uh, what refresh research?
Speaker A
Where is it finding that skill? Is there another, huh? Well, I don't know where it found that skill, but it's using this skill now, directory strategy, and then it's using the researcher sub agent.
Speaker A
So in here, okay. There's refresh research. Oh okay. It was the forward slash command.
Speaker A
Okay. So forward slash refresh research command. It loaded that as a skill basically to get the pipeline and it spawns the researcher agent and that's what we see here, the researcher agent.
Speaker A
And let's see, do we have a researcher agent in here? Yeah. Research RMD cash first demand competitive research for sovereignty.
Speaker A
Alice content map a target to PA questions volumes and competitive signal runs live for data SEO.
Speaker A
You produce a research brief. Okay. So it's going. So here we see, we've got it running.
Speaker A
We're five and a half minutes in now. If I scroll up a little bit, we'll see we had the researcher go.
Speaker A
We had the SEO, AEO auditor go, which is phase B, then the main. And then that agent gave the report to the orchestrator agent.
Speaker A
The orchestrator reads it to decide what's next. The report is good. So let's set up task tracking and see the first batch.
Speaker A
And now it looks at the commands and what's necessary to run. Now I understand the pipeline and it keeps going.
Speaker A
So now we're phase a. Cause we're still in phase a the research brief for the FAQ batch.
Speaker A
And it's just going to keep going. So once again, that's the harness. It's just going to go.
Speaker A
We'll check back in when it's finished. Okay. Boom. Now let's take a look. It's done.
Speaker A
Step one, refresh the research. I didn't spend any money for the SEO stuff. Step two, seed the FAQs.
Speaker A
So new stuff to the database. I think full gate chain half the stuff. I don't even understand, but it understands and it knows why this is necessary and that's what's important here.
Speaker A
Two things you need to know. This is not a clean net new batch committed mid session.
Speaker A
Yeah. That's because in another conversation I was testing it to make sure the system was working.
Speaker A
So yeah, that's okay. Also worth knowing the auditor report only through the things that faxing over there.
Speaker A
Okay. The deploy path change mid session. Yeah. Okay. So I'm just going to explain the issues came because I had a testing everything out and it didn't know you were working.
Speaker A
So that said, make sure everything was ready to go and commit and push and deploy everything.
Speaker A
By the way, that cooked for 22 minutes. Fable did such a better job though.
Speaker A
Didn't need to ask questions. Didn't even need to think. And then, you know what I'll do separately in a new conversation here.
Speaker A
I'll split this. One clod at a recently added section on the landing page and push.
Speaker A
So this should be able to give us an idea of, uh, how, how well it worked.
Speaker A
Okay cool. So we got through it. Now we should see the recently added. Okay great.
Speaker A
This one was submitted with the form a few days ago, the submit a listing.
Speaker A
I had a friend actually submit his community in El Salvador. So now once this is done, we should see new ones here and we'll know that it work.
Speaker A
Meanwhile, I'm just going to do some other little fixes. All right. So it finished the first thing and then I actually set it to write these two more pages that apparently were easy pickings for the SEO.
Speaker A
So it's running the full pipeline on writing these two guides. Okay. So the voice reviewer agent came through plot had to look somewhere else to find that agent and so we're going to have to fix that after and then it found some
Speaker A
problems, unsighted historical claims. So the reviewer is being appropriately strict and adding a source is the honest move.
Speaker A
These are falsifiable claims. Let me revise homesteading cycle one to add a citation. So this is what's cool about the harness that's properly created is it's not just going to write something and make it up to make you happy because then another
Speaker A
agent is going to be like, yo, that agent just made some shit up to make you happy.
Speaker A
Because that reviewer agent's job is to make sure nothing that's not real gets through.
Speaker A
This goes back to why I said I want it to be fully autonomous. I don't want to read it myself, but I want to make sure it's good.
Speaker A
Now, do I recommend you do that? No, I think we should read these things ourselves, but this whole project is kind of more of an experiment for me than, than anything else, but I am going to check it
Speaker A
later. So, I mean, yeah, I mean, to each their own with that homesteading 34 out of 50 ship libertarian 45 out of 50 ship.
Speaker A
Okay cool. So it made it. I had to run through the loop again because it decided that it wasn't good enough.
Speaker A
So I'm just going to finish this up, put it live and we will hopefully see it under the recently added section.
Speaker A
I also added a recently modified section, but nothing can show up there yet because it's like a new field in the database.
Speaker A
Let's take a look. All right. And it's done. We have these new guides. Let's take a look here.
Speaker A
I refresh here on the main page. Okay. I'm not seeing the recently added thing, but let's take a look at these actual pages, the guides.
Speaker A
Yes, they are here. Great. So I just made this posted today. We've got some sources archives.gov.
Speaker A
Okay. Killing the free land myth plainly. Frequently asked questions. And it actually did add these FAQs.
Speaker A
And then let's look at this. What is a libertarian? Somebody calls themselves a returning is not a long, but you're not totally sure when you disagree to.
Speaker A
Once again, a lot of this is written in my voice. Okay. That has my voice standards.
Speaker A
This isn't just AI written. So that's why it looks not like most AI slop.
Speaker A
Then what else it did was it took, for example, the cold card wallet and it added a fax after frequently asked questions and that's all for more SEO stuff.
Speaker A
So it just ran all that through the harness. And it also tells me, oh, there's things we can fix.
Speaker A
So that my friends is a harness that we just built in action with instant results, working on its own for, I don't know, we probably went on over an hour here in and out with the different tasks.
Speaker A
Okay. So we've gone through what an agent to carnuses, how they work, how to get started.
Speaker A
We've built one. So let's TLDR here models intelligence is essentially rented. Perhaps you have a powerful enough computer, but let's just say fable five vanishing three days and disappeared.
Speaker A
Couldn't use it anymore. So we have to keep this in mind. If we want to be sovereign for newers, the harness is the rig around the model.
Speaker A
It's the body for the agent. And this is the part you own open source.
Speaker A
Keep it on your computer. It doesn't matter if intelligence disappears, you can plug different intelligence into it.
Speaker A
Build with frontier, run with open source, save the money on running it, but make sure you get the best quality when you build start by stealing a good one.
Speaker A
You're obviously not really stealing. Shout out to Scott Graham. Thank you for providing this to the world.
Speaker A
It's awesome. Safe agent to work flow. The link is below. You can also just do what I did and just tell cloud code to start building it.
Speaker A
And believe it or not, I'll do a really good job. Swap the brain, keep the body.
Speaker A
That is a durable asset. So I hope this was helpful. Make sure to say hi in the Air Captain's Academy if you want to explore harnesses and rigs, if you have questions. Everyone is welcome. We also host two calls a week in there.
Speaker A
I will be there. We can hang out, ask questions, anything you need. And don't forget to subscribe if you want to learn about AI agents with me. See you later.
Topics:Fable 5agentic harnessAI agent loopAI autonomyLLM harnessCloud Codeopen source AIfrontier AI modelsAI workflowsAI resilience
