Big Idea

The argument in one line.

Prompting is being replaced by loop engineering: instead of manually re-prompting an AI turn by turn, you give it a defined goal, a checklist for what 'done' means, and a budget, then let it attempt, self-check, and retry the work autonomously.

Who This Is For

Read if. Skip if.

READ IF YOU ARE…

You already use Claude Code, ChatGPT, or another agentic AI tool and want to automate multi-step tasks instead of manually re-prompting each step.
You run a business or side project and want recurring AI-run tasks — competitor monitoring, website audits, content research — running in the background without you babysitting them.
You're comfortable writing explicit, checkable success criteria for a task and are willing to let an AI run unsupervised for a defined number of tries.

SKIP IF…

You're looking for tips on wording a better one-off prompt for a single chat response — this is about recurring, self-correcting task loops, not prompt phrasing.
You want a no-code tool; the entire demo runs inside Claude Code's command-line interface.

TL;DR

The full version, fast.

Prompt engineering worked because you had to manually relay instructions and feedback to the AI every turn — you were the loop. New models like Claude Fable 5 are smart enough to evaluate their own output, so the shift is to 'loop engineering': define a goal (the mission), a checklist (what done means), an inspector (a grading pass), and a budget (a cap on tries or tokens), then let the AI attempt, self-correct, and retry on its own. In Claude Code this maps to three commands — /goal for a single mission with a finish line, /loop for a repeating interval, and /schedule for an exact clock time. The video demos both: a one-shot website-audit goal and a recurring YouTube-monitoring loop, then argues you should build trust in autonomous AI in stages rather than granting full autonomy immediately.

Free for members

Chat with this breakdown — free.

Sign in and you get 23 free chat messages on us — ask for the hook, quote a framework, find the exact transcript moment, generate a markdown action plan. Bring your own key when you want unlimited.

Create a free account →

Chapters

Where the time goes.

00:00 – 00:44

01 · Intro

Cold open: prompting is dead, loop engineering is the new master class.

00:44 – 02:44

02 · How An LLM Actually Works

Explains context-window amnesia and why the human had to manually relay every instruction and correction under the old prompting model.

02:44 – 03:36

03 · Models Get Smarter

The new model class (Claude Fable 5, GPT-5.6) needs far less precise prompting because it can infer intent and self-evaluate.

03:36 – 04:43

04 · Looping Explained

The turning point: the human moves from being the middleman relaying feedback to standing outside the loop while the AI self-corrects.

04:43 – 06:13

05 · How To Loop Engineer

The four pieces of a loop: spec/goal, checklist, inspector, and budget.

06:13 – 08:49

06 · The Two Kinds Of Loop

Inner loop (a single self-correcting mission) versus outer loop (a recurring routine that fires inner loops on a schedule).

08:49 – 10:42

07 · Goal Vs Loop

/goal is a single mission with a finish line; /loop is a repeating routine; deliverables must be explicit for a loop to know when to stop.

10:42 – 13:19

08 · Website Audit Demo

Live Claude Code demo: a /goal audits milesdeutscher.com.au like a paid conversion consultant, fetches raw HTML, and writes ranked findings to roast.md.

13:19 – 19:13

09 · Claude Loop Ideas

A YouTube-monitoring /loop demo plus a spreadsheet of ~20 loop ideas across content, marketing, business ops, sales, finance, and personal productivity.

19:13 – 21:20

10 · Scheduling Tasks

/schedule for exact clock times, stopping/cancelling loops, and the Trust Ladder for building autonomy gradually.

21:20 – 21:37

11 · Outro

Subscribe CTA and reminder about the newsletter link for the full guide and loop-ideas list.

Atomic Insights

Lines worth screenshotting.

Loop engineering replaces prompt engineering because new models are smart enough to self-correct without a human manually re-prompting each step.
A loop needs four pieces: a spec/goal, a checklist defining 'done,' an inspector that grades the work, and a budget that caps retries or token spend so it can't run forever.
The inner loop is a single mission with retries — the model tries, checks its own work against the goal, fixes it, and repeats until the finish line is met.
The outer loop is a standing routine that re-fires the inner loop on a schedule, such as every minute, hour, or day.
/goal defines the mission and finish line, /loop defines the repeat interval, and /schedule pins a task to an exact clock time like 6am daily.
Before autonomous loops, the human was the loop — manually prompting, checking the response, and re-prompting in a turn-based cycle.
A well-defined goal specifies exact deliverables (minimum issue count, severity per item, cited page location, ranked impact, a turn cap) rather than vague instructions, because a loop can't know when to stop without a clear finish line.
The demo goal fetched a live website's raw HTML, not just the rendered page, so the audit could catch backend issues like non-functional template placeholders invisible from a screenshot.
A YouTube competitor-monitoring loop can pull a channel's RSS feed on an interval, cross-check it against a running log, and only write a new brief when a genuinely new, topic-overlapping video appears.
Loops can be stopped mid-run with a plain instruction ('stop the loop') or cancelled outright with /loop cancel, without waiting out the budget.
Trust in autonomous AI loops is framed as something to build in stages — turn-based, then goal-based, then time-based, then fully proactive — rather than granting full autonomy immediately.

Takeaway

Autonomous AI work runs on defined goals, not better prompts.

WHAT TO LEARN

The shift from prompting to looping means success now depends on specifying a clear finish line and letting the model self-check, rather than on wording a clever instruction.

02How An LLM Actually Works

An LLM's working memory is effectively one whiteboard per session — once the context window fills, it starts forgetting earlier instructions and quality drops.
Prompt engineering became a skill because, under the old model, the human had to manually re-prompt and correct the AI every single turn.

03Models Get Smarter

The newest model class is smart enough to infer intent from a messy prompt, reducing the need for precisely worded instructions.
Because these models can now judge their own output quality, they can run unsupervised instead of needing a human to catch every error.

04Looping Explained

In the old model, the human sat in the middle relaying every instruction and piece of feedback between the AI and the actual work.
In the new model, the AI completes work, evaluates it against the goal, fixes errors, and retries — the human moves outside the loop and can course-correct rather than micromanage.

05How To Loop Engineer

A loop requires four pieces: the spec/goal, a checklist defining 'done,' a separate inspector role that grades the work, and a budget that caps retries or token spend.
The budget exists specifically to stop runaway loops from silently burning through API tokens or subscription usage.

06The Two Kinds Of Loop

An inner loop is a single mission with retries: it tries, checks itself against the finish line, and stops once the goal is met.
An outer loop is a standing routine that fires an inner loop repeatedly on an interval and keeps running indefinitely.
Missions can be nested inside a routine, so one scheduled check can itself run a full self-correcting inner loop each time it fires.

07Goal Vs Loop

/goal sets one mission with a finish line: the loop stops the moment the condition is true or a turn/time cap is hit.
/loop sets a routine that repeats every interval you define — it has no single finish line, only a repeat cadence.
A goal needs explicit deliverables (minimum count, severity per item, cited location, ranked impact, a turn cap) or the AI can't know when it's actually done.

08Website Audit Demo

A concrete goal example: audit a site like a paid conversion consultant, require severity-rated issues each tied to a specific page location, rank the top by revenue impact, and cap the turns.
The agent fetched the live HTML, not just the rendered page, so its audit could catch backend issues invisible from a screenshot.
The output was written to a shareable markdown file that could then be handed back to the same agent to actually ship the fixes.

09Claude Loop Ideas

A YouTube-monitoring loop can pull a channel's RSS feed, cross-check it against a running log, and only write a new brief when a genuinely new, overlapping-topic video appears.
Reusable loop categories spanning a business include content creation, marketing, business ops, sales/BD, finance, and personal productivity.
Treating an AI agent like a newly onboarded employee — giving it context and explicit cadence — is framed as producing better results than treating it as an ambiguous 'agent.'

10Scheduling Tasks

/schedule pins a task to an exact clock time, distinct from /loop's rolling interval — use it when timing matters more than frequency.
Loops can be manually stopped mid-run or cancelled outright without waiting for the budget to run out.
Trust in autonomous loops is meant to build in stages — turn-based, then goal-based, then time-based, then fully proactive — rather than granting full autonomy immediately.

Glossary

Terms worth knowing.

Loop engineering: Structuring AI work as a self-checking, repeatable process by defining a goal and success criteria, so the model can attempt, evaluate, and retry a task on its own instead of being manually re-prompted at every step.
/goal: A Claude Code command that sets a single mission with explicit finish-line conditions, telling the AI exactly what 'done' looks like for that task.
/loop: A Claude Code command that repeats a given goal at a set interval — for example every minute, hour, or day — until a defined number of iterations or a stop condition is reached.
/schedule: A Claude Code command that runs a goal or loop at a specific clock time, such as every day at 6am, rather than on a rolling interval.
Inner loop: A single mission that retries internally: the model attempts the task, checks its own output against the goal, fixes errors, and repeats until the goal is satisfied.
Outer loop: A standing, recurring routine that re-triggers an inner loop on a schedule, letting a task run indefinitely in the background.
Inspector: A separate model role in a loop that grades completed work against the checklist before marking a task done, acting like a quality-control reviewer.
Budget (in a loop): A cap on how many retries or how many tokens a loop is allowed to spend before it must stop, preventing an unsupervised loop from running away.

Resources

Things they pointed at.

00:00linkAI Edge Newsletter ↗

10:42linkmilesdeutscher.com.au ↗

11:30productMiles High Club newsletter

00:46toolClaude Fable 5 (Claude Code)

00:46toolGPT-5.6 Sol

Quotables

Lines you could clip.

00:46

“So AI is a genius, especially these new models. They're so smart. They're brilliant at every single task.”

punchy framing of the premise→ TikTok hook↗ Tweet quote

03:36

“So before you were in the loop, now you're outside of the loop.”

clean summary line of the whole video's argument→ IG reel cold open↗ Tweet quote

04:43

“Instead of micromanaging the laborer, you can basically just hire a contractor in these new models to do the work for you instead.”

sticky construction-work metaphor→ newsletter pull-quote↗ Tweet quote

14:36

“Just like a real employee, it needs to be properly onboarded.”

reframes AI agents in relatable terms→ TikTok hook↗ Tweet quote

The Script

Word for word.

Read-along

Don't just watch it. Burn it in.

See every word as it's spoken — crank it to 2× and still catch all of it. The same dual-channel trick behind Amazon's Kindle + Audible.

metaphoranalogy

00:00If you're still just prompting AIs like Claude and ChatGPT, you're doing it wrong because the way to use AI now is completely different to what it was just a few months ago. We've moved away from a prompt engineering environment into a loop engineering environment.

00:14And in today's video, I'm gonna give you the master class on how to properly loop engineer. This is exactly how you need to be using AI if you want the best results either in your personal life, for your tasks, or in your business. Instead of prompting AI, waiting for a response, giving feedback, waiting again, giving feedback, waiting again, the looping means that you can run tasks autonomously.

00:36And it also means your outputs are much better because the AI can self correct. But to fully understand how loop engineering works, I need to explain how an LLM, a large language model like Claude or ChachiBT actually works and the very specific change that's happened now that we have the new class of models, Claude Fable five and Charjibouti 5.6 Sol.

00:57So AI is a genius, especially these new models. They're so smart. They're brilliant at every single task.

01:02But the issue is if you're just prompting these models, they essentially have amnesia between every session. They simply forget what you've told it. The context window on chats, whether it's in cord code or whether it's just in a normal chat, tends to fill up, and that's what leads to error messages and low quality responses.

01:18Now in the past, the only way to use AI was really prompting even just a few months ago. You'd have to manually prompt. Even if you had a memory system plugged in, you'd need to keep creating new chats, and you'd need to rely on yourself to keep giving feedback on the AI so it could better complete a task and take feedback into account.

01:38So basically, you would prompt, it answers, you check its prompt, you would then re prompt, and then you'd go around again and again and again. Essentially, you were the loop to get the outcome that you wanted.

01:49You had to control that entire process. And the reason prompt engineering was the skill is because the better your instruction, the better the way you worded the context to the AI, the better its response.

02:01And that's why prompt engineering was a thing. If you consumed lots of AI content through 2024 and 2025 and even the start of this year, it was all about prompt engineering.

02:10How could you prompt an AI to get the response that you wanted? But what's happened in the last few months is that the models have genuinely gotten really smart. The new breed of models, the Mythos class model in Fable five, the OpenAI frontier model in 5.6 Sol, and all future models, they excel in a different area.

02:29And if you've been actually listening like I do to the actual experts, the builders that are actually building Anthropic and have been giving tips on x and giving tips in master classes, the number one theme you'll notice is that they're constantly talking about the loops as the best way to use these models.

02:45Because these models are now actually smart enough to run unsupervised, and they have much better strategic thinking. So you will actually notice if you've used Fable five or if you've used 5.6, when you actually prompt these models, you don't even need to give that much context anymore.

03:01If your memory system is set up correctly and if you use loop and goal, like I'll show you today, you don't really need to have a perfect prompt. In fact, often, I'll give it really messy prompts, but I know I'm engineering it in a way that I'm still gonna get the output that I want. It's because these models are so smart.

03:16They're smart enough to figure out how to deliver what you want reading between the lines. So the real reason we're having this conversation today is really because the tech has changed and how to get the most value out of the tech has shifted. And thus, I do believe prompting as an art form is mostly dead, and it's now moved towards a new model which is looping.

03:38So before it was you in the middle, you had this genius which, you know, kept forgetting things on the left, and then you had the work that you needed to complete on the right, and you were essentially just the middleman between these two things. You had to apply what AI told you to the work, and then you had to relay that and give it feedback.

03:55And honestly, this is probably how you're still using AI. Statistically speaking, most people are still doing this. They're still in the middle asking AI for things, then they're applying it to their work, and then they're going back to AI giving it feedback, and they are basically the loop.

04:07Whereas now, you can have a system with these new models with loop engineering where the AI essentially tries something. It completes the work.

04:17It self corrects, so it's smart enough to know how to critically evaluate the output, and it fixes errors and then tries again. So it actually runs internal loops to complete work.

04:28Once again, this is only possible now because the new models are so smart. So before you were in the loop, now you're outside of the loop. You're able to give feedback.

04:36Of course, you're still able to course correct, and I still actually encourage that. I often do that. But work is actually now happening autonomously.

04:43So now I'm gonna show you how to loop engineer. Instead of micromanaging the laborer, you can basically just hire a contractor in these new models to do the work for you instead.

04:53So, essentially, what happens is you have the contractor. Let's say it's a model like Fable five. It uses a different model as an inspector.

05:01So you could call the smarter model, the orchestrator model. It will, for cheaper tasks, call upon different models. It will review the work.

05:10So it's basically the site manager, in this case, reviewing the work of other laborers. And once it's approved the work, it will mark it as done if it hits the goal that you set out for the AI to do. So you can essentially design these loops, and maybe they take a bit of iterating if you're doing a reoccurring task in your business or in your life.

05:27But once you design these loops, then you can have confidence in the AI to be able to fulfill them autonomously. This is true hands off work, and it's essentially comprised of four pieces. Firstly, you need the spec or the goal of what you want built.

05:39Secondly, you need the checklist. You need to define what done actually means. Then you have the inspector, which the model will automatically orchestrate, and you can tell it also how you would like it to inspect work.

05:51It'll actually go off and grade the work on its own. And then you have the budget. So so the loop doesn't, like, infinitely run.

05:56You can tell it after an amount of tries or an amount of token expenditure where you'd like it to stop. This is how you can stop because I've seen people fall into this trap of running these like infinite loops where you just absolutely burn through your subscriptions or your API tokens. Now I'm gonna give you a demonstration live of how I actually run loops inside Claude code.

06:13But firstly, I want to distinguish if you're confused between the two types of loops because I'm using the word loop overarchingly, but really it's two things. Firstly, you have the inner loop.

06:24This is, for example, if you give the AI a mission like, you know, review my website. Let me know how I can improve. This could technically, even though it's one prompt, constitute as a loop if you use forward slash loop because the AI will orchestrate a set of actions looping internally in order to achieve the goal of reviewing your website.

06:42So it will try something. So, you know, it might scan the website. It'll check if the scan was correct.

06:47If it failed, it would fix it, and then it would try again and it will go through this process. Then it will give feedback. It'll check if its own work actually was good and if the feedback was correct, and then it will try it, and then it will fix it again.

06:58So this is only once again possible because the models are now smart enough to recognize whether their own work is good or not. And you've probably experienced this even just interacting with the new models already. Like, if you ask it for feedback on its own work, it's actually quite self aware, much better than the old models, which frankly were just dumb compared to the new ones.

07:16And then once it's confirmed that it's reached the goal, then it marks it as done. So this is the inner loop. So even single tasks can be defined under the looping model if you use forward slash loop, which I'll show you.

07:28But then you also have the outer loop. This is essentially a job which happens indefinitely, and you can also set it on a schedule, which I'll show you.

07:36So it'll run the job at certain time intervals, it will just keep running, and it will just keep self improving. These are purely autonomous things. For example, if you are on YouTube, you could use a loop on ClaudeCode to scan your competitor channels and make a log of their videos and what's outperforming and the statistics after certain time frames.

07:54And it can just keep updating a document around the clock. Or if you are researching for market alpha, let's say, you know, you're in the stock market or the crypto market, you can use an x scanner every day.

08:04And on ClaudeCode, you can set a loop to run and check for the best viral posts in your niche or or the most, you know, impactful post that might impact your portfolio, which you can feed to Claude, and then it will run the job on an automated schedule. It'll check its own work, fix it, and try again. So you can have an outer loop, which is a reoccurring scheduled task operating with multiple inner loops.

08:26And you don't need to get too confused here, by way. I'll show you the exact way to do it. Chord actually does this on your behalf.

08:31It's all like you need to tell or do an inner loop within an outer loop. No. It automatically does this.

08:35When you use forward slash loop, it can automatically orchestrate. So I'm just trying to explain, you know, how it actually works so you have the the understanding before I show you a real prompt in action. And one last thing before I show you a prompt in action, I need to define what goal is, forward slash goal, and I need to and I need to define what forward slash loop is because we've discussed both those terms in today's video, and I wanna be very clear.

09:00When you use forward slash goal, you're essentially telling Claude in this case, but you can do this on ChatGPT. It works the exact same. They both have this functionality, what you want to be achieved.

09:09So build this thing, and this is when you're done. You give it the exact conditions for when it needs to complete its work. Technically, you know, you can just prompt an LLM and tell it, oh, you know, I want you to do this.

09:21And then technically, that's setting a goal. But by using forward slash goal, you set very defined parameters as to what you want completed exactly.

09:30And this is very important because it's the only way, as I'll show you in a second, that a loop actually knows when to finish. If the goal is not clearly defined, then, you know, how does the loop know when to stop? So forward slash goal tells the LLM what you were trying to achieve.

09:45Forward slash loop shows you the interval in which you were trying to achieve that goal. So run this loop every one minute, one hour, or one day. And this addition that the newer models now have because how much smarter they are, allow you to achieve goals by running automated tasks around the clock.

10:03Or, you know, you could run a loop just for one specific task, and it will simply run an inner loop. So to summarize, forward slash goal is the mission, forward slash loop is the the interval, the routine. And this has been an amazing addition because it's made work so much more autonomous.

10:18And you've probably heard the term AI agents. This is essentially an agentic workflow. If you do what I'm about to show you, you are running AI agents.

10:26You don't need to overcomplicate it. Claude will just spin up agents on your behalf. So by understanding looping and scheduling, I'm also gonna give you a bunch of loop ideas as well to actually show you some examples of what you can do in your, you know, life or your business.

10:38You're actually using AI agents already. You don't need to build an agent. They're built into Claude.

10:42Alright. So I just used forward slash goal. I wanted to scan my website.

10:46I have a website called marsdeutsche.com.au. It's mainly just for our newsletter. It's not, you know, a fully fleshed out personal brand website, but I wanted to audit it.

10:53You know, maybe I could convert better. So I asked her to audit it like a conversion consultant. I'm paying $5,000, write the findings to roast.md.

11:01When it lists at least 10 specific issues, so this is defining the specificity for the goal here. Every issue has a severity, the exact location on the page, and a fix I could ship today.

11:12And the top three are ranked by revenue impact. Every item must reference something actually on the page, no generic advice, or stop after 12 turns. If you don't know how to write a goal from scratch, you could just use AI to write a goal.

11:24You can basically say, look. I want you to review my website. What would be the best prompt for forward slash goal to have a defined set of deliverables?

11:34This isn't as important if you're just asking for a single task like this. But if you're actually running a loop, it's very important to have a defined goal. By the way, this is what my website actually looks like.

11:42It's for the Miles High Club newsletter, basically. It's a very basic website. I'm sure it'll have some feedback.

11:47What it's actually doing is it's reading the HTML. So it can actually look into the site, not just the visuals, but also like the back end. So it can also help you with optimization here.

11:56So because this is forward slash goal, this is essentially just operating like a normal prompt, but with more specificity. I'm just telling it exactly what to do basically. While that's running, let's set up a loop and then we'll check back in on it.

12:09So a loop starts with a goal. So I'm asking it to fetch the latest uploads from my YouTube channel.

12:16For example, if you were a YouTube content creator and you wanted to look at your data from your competitors, so you wanted to see what topics overlap with your niche, their angles, their titles, look for patterns with their thumbnails, you know, to try and spot outliers. You can essentially give it the channel and you can run a loop so you can tell it what intervals to look at the channel.

12:37So you might do this daily, but for the example of this video, I'm gonna show you on the one minute time frame what this looks like on my channel, which is obviously AI Edge. So you can see that it's setting up a one minute loop and running the first iteration now. So I've given it the loop.

12:52So I've given it the loop, which is the interval and the goal, what I want it to do. Now just imagine the power of this. For example, here are some loop ideas.

13:00And if you do want access to these ideas as well as the entire guide from today's video, all of the assets from this video. So you have a full understanding of looping. You can use the link in the description down below.

13:10Become a part of the AI Edge newsletter. We post a lot of really cool stuff there. We work really hard on our newsletter, and you'll also get access to everything from today's video for free.

13:18But let's talk through some ideas while these are running. If you're a content creator, you could run a loop for looking at data and then coming up with ideation for new content every day. Notice if you're a content creator.

13:28If you have any product, you need to create content. So I think distribution is the key in the modern era. So this pretty much overlaps with everything.

13:35All of these goals can be run on intervals. Remember? So if you have a goal of writing YouTube title options, you can also run it with a loop.

13:42So the AI reviews its own work and then you can run an outer loop, basically scheduling an interval for a task to be completed. I'll show you in a second what to do when your computer's off because obviously, you know, if your computer's on, looping will work. But if your computer's off, you'll need to run schedule.

13:56So you can still run loops on a schedule, which I'll show you in a second. But if you're marketing, you can actually run scraping on people's ads or, you know, content that's working well. This is really cool.

14:05If you're running a business and you have the data, for example, let's say Slack, you can put an agent into your Slack. By the way, subscribe because I'm doing a video on exactly how to do that. And then you can run a loop which reads all of the updates from your team channels and automatically compiles a a list of blockers in the business and a summary of the business.

14:22So that's how you can use CordCo to actually audit your business. It can run a loop to audit it, which is I think like a really, really amazing use case of a loop. Now you might think, oh, but I'm not running a business.

14:32You know, you can just run a personal business. You know, you are a business. Any side hustle you have, whether it is, you know, content creation, whether it's digital products, whatever whether it's econ, you you're running your own business.

14:44So why not set up an environment? It can be Telegram, Discord, or even Slack, which I'll do a video on, to run your agents like real employees. You can even give your employees names and run it like a real company.

14:53I I feel like that formality actually helps people get better results out of their agents because it changes their thinking around agents. Because most people have this word like, you know, they see the word agents. They're like, oh, that's really, you know, ambiguous.

15:04But in reality, an agent's just like an employee. Just like a real employee, it needs to be properly onboarded. It needs to be given and empowered with information so it can act.

15:13If you have an employee and you onboard them into your business and you don't give them any information about the business, no context about the business, how do you expect it to perform? If you don't define what you want it to do, in other words, forward slash goal, how do you expect an agent to perform just like you wouldn't expect an employee to perform?

15:28If you don't tell an employee how often you want something done, how do you expect them to complete that task on that interval? So I I feel like people have like this idea of agents and AI as if it's so different from the real world, but it's really not. They're really just digital employees.

15:43And if you treat them more like real digital employees, then you'll just get better outcomes across your agentic work in general. Alright.

15:49Now let's flick back to our ClaudeCode session. Alright. So it's done.

15:53It audited my website. It basically summarized it. So firstly, there's no identity or value prop.

15:58I agree. Through the Zero Social Proof, I agree. We're actually working on a new website to add that.

16:03Three, we have live template pace holders. So it's actually, you you know, seen that if you click through, they don't even actually work, which is true because it can scan the HTML. So this is all really good feedback.

16:15And then it created an MD, so a markdown file of all the things we can actually ship today. And the best thing about ClaudeCode is now I could just ask it to ship these things. Of course, then I'd need to connect, you know, Versal or the domain my website is being hosted on the provider, but it can actually do things on your behalf.

16:30So when you start thinking about AI like this, you know, defining it, giving it a goal, and then empowering it, so giving it the ability to change things, it can like just overhaul your entire business overnight. And we're constantly running loops in our business. I'll actually show you my Slack environment in a future video.

16:46It's very, very cool to just achieve stuff essentially while I'm sleeping. So that was the first goal we set. For the second demo that we're looking at, basically, where it's scanning my YouTube channel to see results, it's now running a one minute looping process.

17:00Some other really cool use cases would be x. Let's say you're training a writing skill because you want better copywriting for whatever you're doing. You could actually look at an account, let's say an account like mine, and you could basically scrape the account to try and replicate the writing style.

17:15And then you can use that to create a skill. Loops or any prompt include a a much better with not only context or a memory system, but also with with skills. So you can have specific skills for specific things.

17:26For example, I have a video production skill. I have a writing skill that knows how to write like me. So you could use loop to automatically scan different profiles, and then you could get it as part of the goal.

17:38So in forward slash goal, you'd say every single time you run a loop, update the skill based on your findings, and then you're actually automatically creating a skill which can empower work in other areas. If you have employees, you can then give that skill to your employees to work on your behalf. You see how how powerful this can be as a business owner.

17:55And by the way, it's not just for business owners. It's also for your personal life. Let's say you have bank statements.

18:00You can run a loop to look at your bank statements. Of course, you know, it's still manual with the bank portal, so you might need to download it and get a CSV file, but there are all sorts of things you can do in your own life. You could run a loop for Claude every hour, let's say, or every day or whatever you want to look through your downloads folder or your desktop and just clean it up in the categorization that you tell it with forward slash goal.

18:20So there are probably all sorts of things in your personal life that you could clean up and audit that that that you can also run on Claude alongside your business stuff. And I'm using it in both areas of my life.

18:31You guys know I built a personal OS. I have my habit tracker. I have all of these, like, gamification things that I've done, you know, bank statements, all that stuff with Claude code.

18:39But then I also have, of course, the business applications, which are, you know, more content and product focus because that's the areas those are areas I'm focused on alongside my my agency. So you can see the loops completed.

18:50It's just meandering while it's waiting for the loop to fire again in one minute. And you can see the loop counter is out of eight because I've defined how many loops I wanted to complete. And you can see if you click on my niche MD, it's it's created a template, which at the end of the process, it will update where it, you know, scan the videos on the feed, the audience, the topics, etcetera.

19:09If you wanna stop the loop, you can just say stop the loop like this and it will stop it. What's very cool is scheduling. So if you want to schedule at a particular time every day, you can also use forward slash schedule every day at 6AM.

19:21Because the goal covers the mission, the loop covers the interval, but the schedule covers the exact times if you wanna do things on the exact times.

19:31If you wanna cancel loop, by the way, you can always go, like, forward slash loop cancel as well. What I actually recommend that you do, because I'll give you the generic list, because I'll you the list that I created using the link in the description below.

19:42But if you really want to get the most out of loops in your business or your personal life, just have a chat with Claude. Just get a Claude Fable five chat or even GPT if you're on GPT and just explain your lives, your goals, where you'd like to use AI, where you're not sure that you can use AI. Just do a one hour brainstorming session.

19:58Do it this weekend. Do a brainstorming session and then ask it. Hey.

20:03Out of all of the tasks that I do, out of everything we discussed, what are the highest leverage things that I could automate? And then it can start telling you, and then you can just go into Claude and apply the exact same steps that we've talked about in today's video. And that's how you go from, like, you know, a turn based system where you're just always prompting and you're always in the middle to becoming essentially automated with AI and having proactive AI that's able to self audit and complete work on your behalf.

20:29I think the natural follow-up to this video is gonna be the agents video where I show you how to use it inside a real business environment because you can actually you know, people think of Slack as some business tool or Discord as some, like, chat or gaming tool, but these are simply applications that you can use to plug in agents to run your personal life.

20:44So you could actually have a Slack, right, with a channel for personal finances, a channel for your habits, a channel for whatever you want, or you could vibe create a dashboard for it, but you get the concept. And then you can have agent employees doing jobs under each thing.

20:57So you might have a, let's say, like a health channel in Slack. You then you might have an agent under the health channel, which is your personal trainer, your nutritionist. And then you might have, you know, a finances, CFO, an adviser, and you can run loops within that.

21:11So you start to see the power of this if you actually treat your life like a company. There's so many cool applications of AI, man. It's super, super exciting.

21:18What a world we live in. But that's it for today. Now you understand looping.

21:21Make sure to subscribe to the channel for more. Link in the description below if you wanna get access to the full guide from today's video with all of the graphics that I showed and the examples that I also showed on the spreadsheet. I think there are some really cool ones in there.

21:33And, yeah, I'll see you in the next video. Have a lovely rest of your day. Peace out.

The Hook

The bait, then the rug-pull.

Prompting, the creator argues, is basically dead — not because AI got worse at following instructions, but because the newest models got smart enough to stop needing a human in the loop at all. What replaces it is a set of Claude Code commands (/goal, /loop, /schedule) that let an AI attempt a task, grade its own work, and retry until it hits a defined finish line.

Frameworks

Named ideas worth stealing.

04:43list

The Four Pieces of a Loop

The Spec (goal)
The Checklist (what done means)
The Inspector (grades the work)
The Budget (caps tries/tokens)

The minimum components required for an AI loop to run autonomously and know when to stop.

Steal forany recurring AI task definition — content research, audits, monitoring

06:13concept

Inner Loop vs Outer Loop

Inner loop: a single mission with retries, stops itself at the finish line
Outer loop: a standing routine that re-fires missions on an interval

Distinguishes a one-shot self-correcting task from a recurring scheduled routine; outer loops can contain inner loops.

Steal fordeciding whether a task needs /goal alone or /goal wrapped in /loop

08:49concept

/goal vs /loop vs /schedule

/goal = the mission (until condition true)
/loop = the interval (every X)
/schedule = the exact time (at 6am)

The three Claude Code commands and what each one controls.

Steal forstructuring any recurring Claude Code task

20:24model

The Trust Ladder

1. Turn-based
2. Goal-based
3. Time-based
4. Proactive

A four-stage progression for how much autonomy to hand an AI agent, likened to onboarding a new employee rather than handing over the company card day one.

Steal forrolling out autonomous agents in a business without over-trusting them too fast

CTA Breakdown