12:10Mark Kashef · Tutorial
6 Ways Opus 5.5 + GPT-6 Astra Upgrade Your Workflow
A practical breakdown of six ways to run Claude Opus 5.5 and Codex's GPT-6 Astra together instead of picking just one.
September 26thThe argument in one line.
A messy Claude and Codex setup isn't inevitable: a single audit prompt run against your real usage history finds the dead weight, and one shared AGENTS.md file with nested per-project instructions stops it from coming back.
Read if. Skip if.
READ IF YOU ARE…
- You use Claude Code and/or Codex regularly and have accumulated skills, rules, or memory files you no longer fully track.
- You run more than one business, brand, or client and your AI tools sometimes mix up context between them.
- You've noticed Claude or Codex ignoring your instructions, repeating mistakes, or burning through context faster than it used to.
SKIP IF…
- You're brand new to AI coding tools and haven't built up any custom instructions, skills, or memory files yet.
- You only use one AI tool on one simple project with nothing accumulated to audit.
The full version, fast.
Claude and Codex setups rot the same way: skills pile up from different tutorials, memory files duplicate and conflict, and switching between AI providers leaves behind rules nobody remembers writing. The fix starts with an audit prompt that has the AI write a script scanning your actual session logs for corrections, dead files, and conflicting instructions, backed by file-and-line citations instead of vague advice. From there, since Claude and Codex both read AGENTS.md, one shared instruction file replaces duplicated per-tool rules. For multiple businesses or clients, a top-level AGENTS.md routes to nested, topic-specific instruction files that only load in context. Schedule the audit to repeat, and filter every new skill through one question: would you do worse without it?
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 →Where the time goes.
00:00 – 02:58
01 · Why your AI setup becomes a mess
Bouncing between Claude, Codex, and trending repos leaves behind duplicate AGENTS.md/memory files, unclear sources of truth, and a stale Obsidian vault that blows the context window before a prompt is even sent.
02:58 – 04:56
02 · The audit prompt for Claude and Codex
Rather than writing a giant audit prompt by hand, describe what to check and let the AI write a script that scans real session logs for corrections, unused tools, and conflicting rules.
04:56 – 05:55
03 · What the audit finds
A real audit of 7,380 messages and 2,518 sessions surfaces hard numbers: 250 corrections, 165 limit-outs, eight dead sound hooks, and skills never opened in either tool.
05:55 – 07:59
04 · Schedule a recurring cleanup
Turn the one-time audit into a scheduled task or slash-command job, and let the AI auto-remediate low-risk items like files unused for seven months.
07:59 – 10:08
05 · The habit that keeps adding clutter
FOMO-driven 'install this repo' and 'paste this prompt' content rarely solves a real problem; most of it can be referenced for one task instead of installed permanently, the way one degree off course compounds over a long voyage.
10:08 – 10:47
06 · Share instructions across your AI tools
Claude and Codex both reading AGENTS.md closes the last gap that forced separate per-tool instruction files.
10:47 – 14:00
07 · Organize multiple businesses
A founder running several businesses replaces four competing instruction files, 15 half-relevant skills, and a 1,400-note vault with one top-level AGENTS.md that routes to a nested AGENTS.md per topic.
14:00 – 16:40
08 · Separate client context and build playbooks
A launch consultant with one folder per client and no written process records a one-hour brain-dump dictation and converts it into playbook files like launch.md and webinar.md referenced per client.
16:40 – 18:28
09 · Three rules for a useful AI setup
Before adding anything new, ask if it's needed on most tasks, whether it can be emulated with one instruction instead of installed, and whether it should be scoped instead of global.
18:28 – 19:03
10 · Get the audit prompt and cleanup guide
Closes with the free audit prompt, cleanup guide, and a pointer to his paid community for hands-on help.
Lines worth screenshotting.
- Switching AI providers without consolidating instructions leaves behind duplicate memory files and conflicting rules that pile up over months.
- A cluttered AI setup can blow your context window before you send a single prompt, because every stale file loads at session start.
- Instead of hand-writing a giant audit prompt, describe what needs auditing and against what framework, then let the model write the script itself.
- Asking an AI to scan session log files for correction patterns beats manually re-reading hundreds of past conversations.
- One real audit of 7,380 messages across 2,518 sessions surfaced 250 corrections, 165 limit-outs, and dead sound hooks nobody remembered installing.
- Back every audit finding with the specific file and line it came from so recommendations are verifiable instead of vague.
- Turning a one-time audit into a scheduled recurring job keeps a workspace from re-accumulating clutter.
- Most 'install this repo, solve your AI's #1 problem' content doesn't solve a core problem worth permanently adding to your setup.
- A trending repo can usually be referenced for a single task instead of being installed and forgotten in a graveyard of unused skills.
- One degree off course looks negligible at first, but compounds into a completely different destination by the end of a long voyage.
- Claude and Codex both reading AGENTS.md means one shared instruction file can finally replace duplicated per-tool rules.
- A single top-level AGENTS.md can route to a nested AGENTS.md per business or topic, loading context only when work happens in that folder.
- In one audited 1,400-note vault, only a small fraction of notes were still relevant, a share that keeps shrinking the longer a vault sits untouched.
- A folder-per-client structure with no written process invites cross-contamination, where one client's brand voice or offer bleeds into another's output.
- One long dictation session walking through your exact process can convert tacit expertise into structured files an AI can reference on demand.
- A handful of plain reference files like launch.md and webinar.md can replace a pile of overlapping skills the AI has to reconcile.
- Before adding anything new to an AI setup, ask whether you'd genuinely do worse on most tasks without it.
- If a skill only teaches one useful instruction, you can point your AI at that single line instead of installing and maintaining the whole skill.
Build one shared system, not scattered files
WORKSPACE AUDIT
A single audit prompt run against your real conversation history can find the dead skills, conflicting rules, and duplicate instruction files quietly costing you context and consistency.
01Why your AI setup becomes a mess
- Switching between Claude, Codex, and other providers without consolidating instructions leaves behind duplicate memory files and conflicting rules that pile up over time.
- A cluttered AI setup can blow your context window before you send a single prompt, because every stale file still gets loaded at session start.
- Loyalty to a specific AI tool is a trap: go where you're treated best, but clean up every time you switch.
02The audit prompt for Claude and Codex
- Instead of hand-writing a giant audit prompt, describe what needs to be checked and against what reference point, and let the model handle execution.
- Rather than having an agent re-read every past conversation, ask it to write a script that scans your session logs for correction patterns and unused tools.
- A good audit prompt asks for dead weight, missing pieces, outdated or conflicting rules, and mismatches between what's set up in each AI tool.
03What the audit finds
- A real audit of 7,380 messages across 2,518 sessions found 250 corrections and 165 limit-outs, numbers worth knowing about your own usage.
- Dead sound hooks, duplicated rules, and skills never opened in either tool are common findings, not edge cases.
- Ask the audit to score each problem area and back every claim with the specific file and line it came from, so recommendations are verifiable, not vague.
04Schedule a recurring cleanup
- Turn the one-time audit into a recurring scheduled task or slash-command job so the cleanup repeats automatically instead of becoming another one-off chore.
- Let the agent auto-remediate anything low-risk, like deleting files unused for seven months, without needing explicit approval each time.
- Comparing rules side by side across tools exposes direct conflicts, like one tool banning something the other allows by default.
05The habit that keeps adding clutter
- Most 'install this repo' or 'paste this prompt' content promises a limitless fix, but rarely solves a core problem worth permanently adding to your setup.
- You can usually reference a trending repo for a single task instead of installing it permanently and letting it join a graveyard of unused skills.
- Small deviations compound: one degree off course on a long voyage lands you somewhere completely different by the end, same as one bad habit repeated every session.
06Share instructions across your AI tools
- Claude and Codex now both read an AGENTS.md file, which finally lets you maintain one shared instruction file instead of duplicating it per provider.
- Writing rules once instead of forking them per tool removes an entire category of conflicting instructions.
07Organize multiple businesses
- When one person runs multiple lines of business, a single top-level AGENTS.md should act as a router that points to a nested AGENTS.md per topic or business.
- Nested instruction files only load when the AI is working inside that specific folder, which keeps unrelated business context from leaking into every session.
- In one audited 1,400-note vault, only a small fraction of notes were still relevant, a pattern that gets worse the longer a vault sits untouched.
08Separate client context and build playbooks
- A folder-per-client structure with brand docs, call notes, and no written-down process invites cross-contamination, where one client's assets leak into another's output.
- A single long dictation session capturing your exact process step by step converts tacit expertise into structured files the AI can reference on demand.
- Replacing a pile of overlapping skills with a few plain reference files, like launch.md, webinar.md, and emails.md, is often simpler than maintaining a skill.
09Three rules for a useful AI setup
- Before adding anything new, ask if it's needed on most tasks and whether you'd genuinely do worse without it, rather than installing it just because it's trending.
- If a skill only teaches one useful instruction, consider pointing your AI at that single line instead of downloading and maintaining the whole skill.
- Scope anything that's only occasionally useful instead of loading it globally, and treat maintenance as a recurring job, not a one-time cleanup.
Terms worth knowing.
- AGENTS.md
- A shared instruction file now read by both Claude Code and Codex (and Gemini), letting one file replace duplicated per-tool rules.
- CLAUDE.md
- Claude Code's dedicated instruction file. Running it alongside a separate AGENTS.md invites duplicated or conflicting rules between tools.
- AI operating system (AIOS)
- The full stack of rules, skills, memory files, and automations a person builds around their AI coding tools to run their work.
- Context window
- The amount of text an AI model can hold in a single session before it starts dropping or deprioritizing earlier information.
- Skill
- A packaged set of instructions, often downloaded from a tutorial or repo, that an AI tool loads to handle a specific recurring task.
- Scheduled task
- An automation that re-runs a prompt, such as a cleanup audit, on a recurring interval without a manual trigger.
- Sub-agent
- A specialized AI agent spun up to handle one narrow part of a larger task, such as reading a provider's official documentation.
Things they pointed at.
02:30toolObsidian
14:30toolClaude Code Guide Agent (sub-agent that reads Anthropic docs)
14:30toolOpenAI docs skill (for Codex)
Lines you could clip.
00:00
“If you've been building or using an AI operating system for a while now, then odds are it's a mess. And it's likely not your fault either.”
01:29
“You end up having an empire of files and you basically have an AI operating system that is occupied by the AI slot monster.”
08:27
“This repo will solve Claude Codex's Gemini's number one problem. And when you actually look at it, it's not really solving a core problem.”
09:14
“Imagine you're sailing from west to east... you changed your bearings, you went one degree in the wrong direction. One degree on its own might seem like absolutely nothing. But by the end of it, you'll be somewhere completely different.”
17:45
“You'll have an increasingly more powerful dragon that doesn't need as much in terms of leash.”
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.
metaphoranalogystory
If you've been building or using an AI operating system for a while now, then odds are it's a mess. And it's likely not your fault either. Over the past year, you might have gone from lovable to cloud to codex and then back to cloud.
And along the way, you installed every skill, every top trending GitHub repo. You let it all accumulate and now you have a pile of assets that don't work together. And on top of all this, even when you use frontier models, they still ignore your rules, say they're done when they're actually not, And most importantly, burn through your entire token limits in the snap of a finger.
So in this video, I'm going to show you how you can take your workspace of chaos and bring order to it. All right, so just a few months ago, AI operating systems, or the term AIOS, became all the rage. But after I've been on dozens of calls with founders, I keep seeing the same pattern over and over again.
So here's an example of a hypothetical but realistic version of an operating system that I see all the time. And this is the clean version, where you'll have a series of agent MD files, a series of memory files, one called memory number one, number two, final, then the real memory. And then you'll have a hodgepodge of different markdown files, a sole .md that used to be used for an open claw or a Hermes, and then everything lives in one melting pot.
And the real crux of the issue is that you end up having a bunch of half -built bridges where you built a content operating system, a launch operating system, a growth operating system. And we'll go through some actual examples that I've seen on calls with actual founders and what they're looking to build. But because of the sheer volume of the number of things that get dropped every single week, the endless GitHub repos that go on trending, that grab your attention away from what actually matters and makes you money, you end up having an empire of files and you basically have an AI operating system that is occupied by the .
And one of the biggest culprits for this is this constant migration from model provider to model provider, where in general, most of us are very fickle. I have no loyalties to Claude, Codex, and even if Gemini wakes up one day and is actually a passable model, and I move over there, I'll always go where I'm treated best, and I'm sure many of you are the same.
But the cost of all of the junk that you have on your computer as a result of running tests, building things without actually looking at what it's doing, having all of these temporary versions of files running on your computer that never get deleted, is you have a very bloated infrastructure. And you might have something like an Obsidian from months ago, back when, again, it was very sexy and it was all the rage, and now you let it sit idle.
And a lot of the files within Obsidian are expiring or are expired or are stale. And then you end up with hundreds, if not thousands of files that live outside of Cloud and Codex on top of the memory files and all the other junk that you have within those actual folders themselves. And then when you combine this all together, you blow your context window before you even send a single prompt.
But realistically, we should be able to go from something like this to something like this just by using prompts and auditing what you already have. Now, this is the part of the video where you would expect me to give you some huge comprehensive prompt to run your audit. But realistically, you don't need that because we do have models where all you have to do is explain exactly what needs to be audited and against what reference point and what framework and it'll take care of it for you.
All you'd really have to say is something like this. I switch between Claude and Codex all the time. Look through my setup in both and tell me what's working, what's dead weight, what's missing, and that includes my skills, rules, instruction files, settings, plugins, connected tools, agents, and every project that I run as its own system.
Now this is the section that I want you to pay attention to because it is a bit more nuanced than you might have seen before. Typically, I would ask Claude or Codex to spin up a team of sub -agents to go through all the conversations we've had in the past to actually read them end -to -end and analyze all the tools I used, all the difficulties that I had, etc.
But you could have hundreds, if not thousands of conversations where this would add up pretty quickly. So instead of that, I actually ask it to write a quick script. And this script can look for certain keywords that are very adjacent to something like me pushing back on Claude, me frustrated with Claude.
or it can actually go through the entire JSON -L files, and I'll show what this might look like on screen, where you can pinpoint certain tools being called or not called, and you can pinpoint that through a script. And then I tell it to tell me what I rarely or never use, which rules are outdated, what I have set up in one tool but not the other, especially if you want to be more model agnostic and be able to be nimble enough to hop from Clog to Codex to insert name of model here.
In number four and five, go through things like what settings, automation, scheduled tasks, or elements of my project seem to be holding me back versus pushing me forward. And instead of you acting like you're reading a markdown file, you can ask it to create it as a webpage so you have that visual element to go through it and actually make sure that whatever it's telling you to remove or what recommendations it's giving you actually makes sense.
So around 30 minutes later, we get a report that looks like this, and it goes through all the messages that I've typed, the top 10 things that you want to change the most, and some things I'm surprised by looking at them right now. So I have eight dead sound hooks. This is from way back in the day when I used to use cloud code when it first came out, and I wanted it to make a pinging sound when it was done because there was no remote control.
There was no additional way to do that through notifications. So you never know what might be hidden from a prehistoric time of using these tools. Now, based on going through all the conversations between both Codex and Claude, it basically said that I should put all of my writing rules, all of my removal of AI -isms, slop -isms, all in one shared files that both tools can load.
And then as you go down the list, it goes through things where it tells you have a bunch of random junk files. And these are probably taking gigabytes, if not terabytes of data on your computer. And then it will score each area so you can see exactly where your problem lies.
And I promise you, if you do this and then you do a schedule, it could be a schedule task in Codex or a slash schedule in Cloud, and you have it run this report and ideally remediate the majority of things that don't need your consent. Let's say it's removing dead files that you haven't used in seven months, assuming they're not super important.
And you'll have this self -improving or at least. self -cleaning system that doesn't end up bloating over time or accumulating a lot of dust. And then as we go down, we can see every single rarely used skill, many of which you'll see here are Obsidian because I stopped using Obsidian a while ago.
If you run something like Slash Insight, which is a older mechanism that you can use to reflect on all of your conversations, it's helpful, but it's not as helpful at telling you what to delete versus what to add. And when it comes to rules, especially if you're less technical or you don't have as much time to invest in Claude code, and you've been watching videos, taking the transcript, telling Claude, go do this.
And it's creating these skills or rules for you. You'll notice that you'll have rules that conflict between Claude and Codex. So if you end up taking a task and offloading it for whatever reason, it might not execute the way you expect.
And all of the rest here are things that you can quickly scan. And for the most part, you should be at peace with giving it the ability to go through and prune your entire systems so you can remove a lot of the additional information and noise that's causing the issues you're probably facing. Now, one cool thing that's been added to the Claude desktop app is after you run this review using my prompt, you'll get a recommendation to run the low risk configuration cleanups.
Now you could take a look at it and say, you know what, go ahead with everything except for X, Y, and Z, but you can run this right away. And nine times out of 10, when I've ran this testing the prompt, it gives you things that don't have a very big blast radius. And one important thing is if you take a look at context, even with scanning both environments and all the rules and skills, just because we did it in a slightly more efficient way, we're only using a third of our current context window.
Now, that was a way to get things more efficient in one shot. But this problem will come back if you fall for the exact same traps every single time. Now, it's one thing to be able to clean your workspace, and it's another one to clean your habits.
If you are someone that falls for FOMO and always feels like they need to know what is this 1 % change in the way that skills work that I have to watch this YouTube video and update and tweak my skills. This is for you.
You'll see things like stop doing this and start doing that. Paste this one thing into Claude and you'll never run into limits again. And it will also solve world hunger and world peace.
Or my least favorite and the one that most people fall for is This repo will solve Claude Codex's Gemini's number one problem. And when you actually look at it, it's not really solving a core problem, or you don't need to install it.
You could literally run an entire job. You could make it a part of the prompt where you say, go and refer to this repo for this specific task. But you don't necessarily have to download that skill and add it to your heap of other skills that are pretty much in a graveyard.
And you end up looking something like this, where you stack a bunch of skills that individually are very helpful, but in aggregate, one, they overlap. Two, they actually have mixed signals. And three, you probably don't update them to the latest versions.
So over time, you forget about them. And as you forget about them, they start steering your agents in the wrong way ever so slightly. Now, let me give you an analogy to quickly drive this home.
Imagine you're sailing from west to east and you're going from somewhere like New York City all the way to Glasgow. And for whatever reason, you changed your bearings. You went one degree in the wrong direction.
One degree on its own might seem like absolutely nothing. But by the end of it, you'll be somewhere completely different. So although it's tempting to constantly give in to the latest and greatest and staying on top of every new model, every new framework, every new skill, every new technique, every new prompting change.
If that comes at the cost of you maintaining and upkeeping what you've built, then you might not drift by just one degree. You might drift by five, 10, 15 degrees. And you'll notice that in every single session.
And you'll spend a lot of time and tokens in that process, not making any progress at all and just spinning in circles. So instead of paying attention to this kind of stuff, what you should pay attention to are the slight nuance changes that happen that make it that much easier to be nimble and model agnostic. A really good example is when Anthropic finally made it so that Claude can respect and look for an AgentsMD versus a ClaudeMD.
Up until now, if you wanted to work in Claude and Codex, you'd need a ClaudeMD and an AgentsMD. Now, Gemini also uses a version of AgentsMD, so an AgentsMD would work for both of those providers, but you were missing that final piece. Now, you could write one AgentsMD.
one set of skills, one set of rules, and find a way to have it apply to all of them. Now to give this some more color, I'm going to walk you through two actual examples of founders that I've recently sat with and what their AI operating system looked like before and after. So this first person, let's call her Sarah.
And Sarah runs not just one, but multiple businesses where each business has its own set of voices, its own set of content, and its own set of projects that need to be executed. So what she was suffering from is something like this, where she went from Claude to Gemini, back to Claude, to Codex, then back to Claude again.
And you end up having a ClaudeMD, an AgentsMD, a Gemini, and then in between there was a quick transition to HermesAgent, which she kind of reeled back and went back to the normal model providers. So the first core problem is you have all of these different files competing to give instructions as soon as you open your brand new session in Claude, Codex, Gemini, etc.
And then we have 15 skills of here. Only three of them were actually relevant in driving the business forward. A lot of the rest were just things that were recommended on YouTube, and she felt the need to download them because they seemed great on their own, but she didn't actually take it, create a derivative of it that fits perfectly into her stack.
And then on top of that, she has an obsidian. And in that obsidian are 1 ,400 notes. Around 10 % of them are actually relevant to today's work.
And you can imagine in the future, this will rot to 5%, to 1%, and so on. And last but not least, we had a global Cloud MD. And if you don't know what this is, a Cloud MD obviously will give you instructions that are loaded at the runtime of starting a session.
If you make it global, then every single runtime, irrespective of the project, will inherit a lot of those instructions that you would have to manually override as long as you remember them. The issue is when you don't actually remember them. So this causes things like mixed signals, rules that AI actually doesn't listen to, and a whole host of other problems.
Now the after looks something like this, where we pruned and went through all the different files to create one unified agents MD. And realistically, it looks like this, where you have a Venn diagram, the one with three different circles. And once you add the last circle here, assuming you only use three providers, you're going to have things that are in common between all of them.
Maybe a skill, a set of rules, instructions that you find an easy way to check off. We'll have other ones that are specific to that specific workspace. And when you were working with that provider, some of those are stale, some of them are relevant.
And the goal, again, is to unify them. so that we don't have anything that's specific to a provider unless that paradigm, that structure is only available with that provider. And simplifying the rest of the operating system is straightforward.
You just create different folders that contain the different parts of the business, the lifestyle, etc. And within those, you can also have a nested agents MD, where this is not loaded when you start a brand new session. You only load this agents MD on the outside or some global agents MD on the outside.
And when it comes to optimizing this operating system for all these businesses, it could be as simple as creating one individual folder. per line of business. So you theoretically use this agent's MD as your air traffic control, your pointer, and you would use it to point at each individual folder.
And once you're in that folder, whatever language model you're using will read this next agent's MD that's specific to what we wanna do for fitness in terms of the content, the structure, the script writing, the video editing style, et cetera. Now I'll take another founder that I worked with and let's call her Beth. And Beth has a completely different line of business.
She works in launching and marketing campaigns. Beth is an expert at creating go high level funnels, web pages. She runs cold traffic webinars.
She's great at Facebook ads and the whole gamut. And this is what her operating system looked like. She had one folder for every client that she had.
And for each client, some of them might have skills that are specific to them. Some might have agents MD and some might have absolutely nothing. And after admitting to watching tens of YouTube tutorials, she had two second brains and she was using neither of them.
So she had to depend on her actual one. And I would say the worst offense is she had so much tacit knowledge about what it is that makes her great at her profession. And she didn't take the time to brain dump all of that and have Claude and Codex use, let's say, the Claude Code Guide Agent, which is a sub -agent that specializes in reading the anthropic documentation to tailor and...
help you massage your workspace to work perfectly with cloud models. And the same thing with the opening eye doc skill, where if you're using Codex, it can go read up on the latest and greatest best practices. So when you tell it, here's my business, here's what I do, and here's my goal, it can take all of that to help you marshal a good plan to optimize your workspace accordingly.
So the result was a crossing of wires where many times the launch assets for client A would accidentally seep into client B and vice versa. And this is what the after looks like, where once again, we have one unified agents MD. We have a new folder that's called playbooks, where we took a full brain dump.
We had a one hour dictation where she went through the exact steps of how do I start, ideate, create the first version of a funnel? What does it look like to get someone on a cold webinar? What do we do on that cold webinar?
Imagine mapping out every step, but doing it just through a dictation. We then were able to create two simple files like launch MD and webinar .md and emails .md. And instead of creating skills, we could have the AI reference the specific files anytime that there's a task where it's pertinent to check it.
And when it comes to clients, we just created a simple clients folder that had client A, B, C, et cetera, and each one. just like before, could have another nested agents MD that gives more information to contextualize what we're doing for this client, how it differs from others, and the specific skills, rules, et cetera, that are pertinent to those specific domains.
So hopefully those two basic examples show you how elegant and simple the solution could be to taking a workspace that is holding you back to one that allows you to do what you need to do and focus on working on the business versus working in the business. Now, one thing I'll leave you with is this three -step framework that you can use to help protect your existing workspace and ideally your mindset to intentionally have tunnel vision and only focus on what matters.
When you see that next model or next repo, instead of immediately pulling for the link and downloading it and then using it for two days like a shiny toy and then throwing it into your toy box so you never revisit, you can do this. When you see something new and you want to import it into your sacred space, you can ask, is this thing going to be needed or helpful on every task or most of my tasks?
If you didn't have it, would you do what you're doing today worse? Which would beg the question typically, is there a way you can emulate it? Can you go from using a skill and downloading it to literally pointing your AI at the skill and saying, hey, there's one thing in here that might make you work better on X.
Can you scan the skill and just update the one instruction that might be that additional alpha? Sometimes a lot of these skills all boil down to one simple line of instruction that you could use to take your AI to the next level. The next is when it comes to scope.
Is this only useful sometimes? If so, it could make sense to load it certain times versus others. So it might not be a global skill.
It might be a very specific skill, rule, or asset that is pulled once in a blue moon. And last is really the maintenance piece, which is kind of running that scheduled job using the prompt I showed you earlier in the video. As we get better and better models, I can guarantee you will need less skills and less rules.
You'll have an . to keep it controlled. And if you know that, the job is to be proactive about it.
So you can actually focus on getting the incremental benefit of these better models versus always having to fight with Claude, Codex, whatever language model of choice, and not actually making the progress you're looking to make. And obviously I'm gonna give you that prompt I showed you along with a little guide that you can feed your AI of choice to help it help you better decide what is worth cleaning, retiring, and adding.
And as always, if you like content like this and you want access to much more and access to me where I'll send you a personalized loom in my community to help you with your specific problem, make sure to check out the first link down below for my early AI adopters community. This is literally what I do all the time and I'm an AI consultant, a professor, and this is all I eat and breathe.
For the rest of you, if you found this helpful, I'd super appreciate a like and a comment on the video and I'll see you all in the next one.
The Hook
The bait, then the rug-pull.
Mark Kashef opens by naming the quiet failure mode of every AI power user: months of switching between Claude, Codex, and whatever's trending leaves behind a pile of duplicate memory files, dead skills, and rules that quietly contradict each other. The video is his walkthrough of how to audit that mess with your own usage history, then rebuild around one shared instruction file.
Named ideas worth stealing.
16:40list
Three-question filter before adding anything new
- Is this needed or helpful on most tasks? Would you do worse today without it?
- Could you emulate it by pointing your AI at the one useful instruction instead of installing the whole thing?
- Is this only useful sometimes? If so, scope it to specific tasks instead of loading it globally.
A three-step gut check for deciding whether a new skill, repo, or prompt deserves a permanent place in your setup, versus a one-time reference or no install at all.
Steal fordeciding whether to install any new Claude Code skill, MCP server, or trending repo
10:47model
content-os/ AGENTS.md router
- One top-level AGENTS.md as the router
- One nested AGENTS.md per business or topic folder
- A shared skills/ folder for cross-cutting tools
- An archive/ folder for retired notes and old rules
A single top-level instruction file points each topic or business to its own nested AGENTS.md, which only loads when the AI is working inside that specific folder.
Steal forany multi-project or multi-brand folder structure that needs isolated context per line of business
14:00model
launch-os/ playbook structure
- One top-level AGENTS.md (30 lines)
- playbooks/ folder with launch.md, webinar.md, emails.md
- clients/ folder with one subfolder per client, each with its own AGENTS.md
- archive/ for old call notes and retired second brains
A consultant replaces per-client skills and scattered call notes with a shared playbook folder plus one isolated, nested AGENTS.md per client.
Steal forservice businesses that need client work kept from bleeding into other clients' assets
How they asked for the click.
VERBAL ASK
18:28product
“if you like content like this and you want access to much more and access to me where I'll send you a personalized loom in my community to help you with your specific problem, make sure to check out the first link down below for my early AI adopters community”
Soft pitch placed after the free audit prompt and cleanup guide are already offered, framed as deeper 1:1 access rather than gated content.
MENTIONED ON CAMERA
FROM THE DESCRIPTION
PRIMARY CTAWhere the creator wants you to go next.
OTHER LINKSAlso linked in the description.
Visual structure at a glance.
