Claude Loops: Build Agents That Run Without You

Today, I'm going to show you five different Claude agent loops that you can build to basically have Claude running without you autonomously to build and automate anything you want. Now, if you're wondering, OK, what are loops? Essentially, this means that you don't need to prompt anymore because you can give loops to your AI agents like we've got right here. And I'll explain exactly how it works in a second.

And each one acts, checks its own work with an independent judge, and repeats until it's actually done. Now, inside my agentic operating system, I've actually built out five different loops, one to build anything to a spec, one that works with a Kanban board, one that takes an idea from inbox to shipped, one that watches the news, and one that writes your SEO, all running on one shared memory that makes them smarter every time. So I'm going to explain exactly how each loop works. Now, the difference here is you are no longer the person quality controlling or checking anything, which saves a lot of time.

And also, these loops can run without you autonomously. Or you can actually add a human in the loop, which I've also built out. So you can have two different styles of loops depending on what you want to build. If you saw the tweet from OpenClaw founder Peter Steinberger recently, he was talking about this as well.

And essentially, a system where you're not going back and forth with AI anymore. This saves so much time. It just runs in the background. And you can get way more done with these powerful systems.

In fact, your only job now is to make sure you're the architect, right? So before, and you probably can relate to this, for a year, I was the loop. So I'd type a prompt, I'd wait, I'd read the answers, spot what was wrong, type again and again, and nothing would move until I was sitting there driving it. So my output was pretty much capped at how long I could stare at a screen.

And for a lot of people watching this, you can probably relate to that. Now what we have is a system where we build the loops ourself. So I define what's done and what that looks like. An agent acts, a separate judge checks it against a goal, and it repeats until its own work passes the quality control checks.

Now we have five of these running inside the agent OS, building, planning, shipping, watching, and writing. And they all run whilst I sleep, and then I wake up to the work done. And you can run the same, right? We've actually got loads of people using the agent OS inside the airprofit.

One, we have over 196 pages of testimonials and wins from people using systems like this. So if I can do it and I'm non-technical, and you might be non-technical too, you know, this is a pretty powerful system. So how does it work? Well, if you look at the old way versus the new way, one makes you the engine, the other makes you the architect.

So for example, with the old way, with the prompting using the old way, you type, you read, you wait, you correct on every turn. So most people are going into chat GPT, for example, like this, and they're just going back and forth. And it takes a lot of time, it's very inefficient. And the problem with that as well is like, nothing moves unless you're at the keyboard.

There's no check on the work, but your own eyes. And the output is capped by your attention. With the new loop systems, you write the loop, it runs itself, you define done once, the agent acts, a judge checks it, it repeats on its own until it passes, and it runs whilst he sleeps, you wake up to it done. And every loop is the same shape, which is act, observe, judge, and repeat.

So it loops round and get stuff done without you. Or you can have like just a quick check to say, okay, this is good to go. So let's talk about the first loop here, which is loop engineering. So for example, if we have a look at this system, we have loop engineering.

And what we can essentially do here is we define what's done, we give it a starting point if we have one, and then we can select the builder, aka the API for the builder, how many rounds it goes round, and then which API we use for the judge. So you could have like a GLAM 5.2, which is a powerful model for the judge. And then you can have a free API for the builder. And that means number one, you use less tokens.

And number two, it costs less. And number three, this just runs round in a loop. So you can say, okay, build out a website. And we've built out several things here, as you can see.

You can say, build out this project, and it'll loop round however many rounds you have. The builder will create it and the judge will judge it. And then it just loops around until the work is done. And I don't need to touch anything.

And also we can just see the outputs we've created over here inside our workspace. So it will get saved and I can check it up later. Next up, we have the Kanban board. So this is loop number two, a loop with three roles instead of one.

So for example, you can have a planner that breaks your goal into cards, a builder building each card, and a reviewer that checks it actually landed before the card moves to done. Let me show you an example of this. So if we take a look at this, we've got this content system where we created a video and a step-by-step guide with a content judge. So we have separate agent profiles.

We just drop our task inside the content board and then we can say, okay, build this video, create this blog, et cetera. And it can actually create the video or it can create the blog post as you can see. And essentially a video builder would create the video, a director would help with that, and a content judge would actually look at the content and see if it's good or not. And then once it's done, it all goes around here.

So automatically the task gets triaged. It goes into triage to do ready running blocks if it can't move forward and then done. And I don't need to touch anything. The loop helps you quality control the work because it's just running on autopilot.

Then we also have the pipeline. So the pipeline system is pretty powerful because you can use this loop for turning ideas into finished work. So you capture an idea, an agent classifies it, routes it and plans it, you approve it once, then a project manager and its sub-agents build it out and the whole thing lives in your vault so nothing gets lost between an idea and it's done. So if we have a look at this, for example, we have a system over here where we can basically go from idea to implementation without any sort of guidance from me.

So if we have a look at this one, for example, this feature for an AI SEO app, we can let the agent shape it. And then what it's going to do from here is ask us to approve it. So we're going to say build and approve. And then from there, it's just going to start building out that example, right?

So we can click on build the deliverable over here and it'll actually get completed, which is pretty powerful stuff. So this is running in the background, it's getting completed. And then once it's done, you'll see all of our done projects over here. And so these are all different apps that were built.

We can see them, preview them. And literally the only thing that we have to do is just approve the plan and we can delete any ideas too if we don't like them. And so the great thing about that is that we literally go from idea to done and all I need to do is just approve it. And I think that's a powerful loop because when you're looking at this, for example, like this system here, you can go from, if you have like lots of ideas, but you just don't have the time to implement them, you can build something like this out and you can see everything that you've built and you can come back to it later and it's pretty easy and simple, right?

So everything is set up inside one system. We can review it. But the main thing is here that I don't need to be the one building. I don't need to be the one checking the work.

I don't need to be the one like giving feedback or prompting or anything like that. So it's super powerful. Then we have the Oracle. Now the Oracle runs on a loop.

This is built with Hermes Agent and essentially it's a loop on a timer. So every morning it searches for the latest breaking news in AI because obviously I need to stay updated with that. It ranks it by relevance and hands you the six biggest stories to post content about today, each with the original source, your unique angle and a ready hook. So one click turns any story into ranked SEO articles on your website.

Also logs every news update to our Obsidian memory. So for example, over here, this is running, you can see it was last consulted four hours ago. We can see that we've published loads of content over here. So for example, like this, this is a article we've already published on that trending news topic and we can do that every day, day by day.

And the great thing about that is that it's just running on a loop. I don't need to prompt it. I don't need to say like, hey, check the latest news or search for me the latest news or give me content ideas. I've got six of them ready to go at any time.

So for example, this one will be a great topic to create content around and it's just running in the background for me. The same with, for example, Hermes Astros. Hermes Astros does the same thing. It monitors certain keywords and then gives us content ideas with unique angles based on those keyword ideas.

And then the fifth loop is the SEO pipeline. So the way this loops work is basically it generates keywords for us, right? So we can research keywords to create content around. Then we can click on use topic.

Then we can generate the articles and we can generate SEO content in like one single click. If you want to see an example of that, you can see all of these articles that we've published right here on our particular topic, super easy and simple to set. And so it's an SEO content loop. Now, the final loop that makes everything compound and work together is the memory system.

So every time we are using these agents, we update our memory automatically. So if we look at the recent, you can see that Hermes Apollo just got updated inside our memory loop and this is updated every few hours automatically. And so this runs on a loop because the memory is automatically updated by our agents and then our agents pull up the context and use that to update me and find new ideas and create better outputs for me, personalize me, because it knows inside my memory galaxy, my projects, who I'm working with, my team, my vision, my voice, my brand, my companies. Everything is connected inside one memory galaxy.

And so that runs on a loop as well. So that's basically it. That is every loop system that we've built. We built it out with Claude and a lot of the systems run in the backend with Claude CLI and Hermes as well.

If you want every loop from me, if you don't want to wire up five different systems, then you get the full AgentOS system inside one dashboard with one shared memory ready to run inside the AI Profit Board. And we also get four weekly coaching calls where you can ask questions, share your screen, get help and support in real time as well. So inside the AI Profit Board in here, you can get access to our AgentOS. And this is our community for helping you learn, save time and grow with AI automation.

The great thing about getting the AgentOS is like some people think you need to be technical to run loops. But if you get our loops and our systems, then you don't need to be technical because you can just give this zip file to Claude or Hermes and it will set out for you. You can see when it was last updated, you can get a video tutorial on how it works. You've got a full build out guide here.

And then we've also got new daily guides based on what's actually useful, as you can see right here. So we update it all the time. We also answer community questions daily and I personally answer them. Plus there's always people online to help you 24 seven.

And inside the calendar, you can jump a week of coaching calls, share your screen, ask questions, meet people in real time. Inside the map, you can meet people in your local area who are building with AI agents like you. And that's all available inside the AI Profit Boarding. Link in the comments description or just go to the AIProfitBoarding.com.

Thanks for watching.