So What? Getting started with Claude Skills

Well, hey everyone, happy Thursday. Welcome to So What, the Marketing Analytics and Insights Live Show. I am Katie, joined by Chris and John. We got the band back together this week, guys. We did.

Kind of a lame half high five, but we'll take it. Um so we're together this week. The next couple of weeks are gonna be up, down, left, right, center. So, you know, we are knee deep in event season. We hope to catch you at some of the live events.

But this week, we are here and we are bringing you the live stream, so what? Getting started with Claude Skills, which I'm actually really excited to learn about because I know Chris, obviously, the second something is released, you're on top of it, you're playing with it, you're testing it. And I was using Claude the other day, and skills started kicking in that I didn't ask for, and I was like, uh, whoa, whoa, wait a second. But that I also couldn't find where the skills were to actually look at them to see what they were comprised of. So I have lots of questions, as expected.

Where should we start? We should probably start with what the heck is this? Uh and and why would anyone care? So for folks who've been using chat GPT for a while, you probably know that you can invoke a GPT within a within a chat, within a new chat by saying, you know, at writing style analysis, or at you know, uh virtual Katie. What do you think about this thing?

And it will can reference that. But those GPTs are somewhat limited, um, and they're not available in other systems and things like that. Um so if you are in say OpenAI's codecs, their coding environment, your GPTs are not there. Um if you're using the API, it's it's very difficult to do that. But that was sort of the the the beginning of this of this concept of if you have a pre-packaged thing, like a mini app, you can call it within a conversation.

Well, about a week ago, Anthropic uh, which is the maker of Claude introduced their version of this called skills, and skills are essentially the ability to take a uh a similar concept, like a writing style analyzer, a business uh requirements document analyzer, whatever the things that you would normally use as a Claude projects, but turn them into skills that you could invoke in a conversation at any time, new chat, old chat, etc. And two things that make it very interesting. One, you can take the skill and you can hand it as a like a plug-in file to another claude user, any other cloud user. They can drop it into theirs and immediately have access to it. And two, it's usable in environments like Claude Code, the coding tool.

So Katie, you look like you have some questions. I do. Um, and it might just be me. Um, you know, so John, I would love to get your perspective. I didn't know, I wasn't aware that in something like an open AI, you could add a custom GPT in a different conversation.

So that's my learning number one today. I did I wasn't aware that that was a thing, um, which is actually pretty cool. So, like, if I'm understanding correctly, so you're in the just what can I help with? You're not in a custom GPT. Right.

So you could start a conversation and almost like you're just tagging a person in Slack or any other social media platform. For example, you can tag a custom model that you've built inside that same system. Sometimes it's uh really unreliable. Um you do the same thing in Gemini? You cannot.

Okay. I'm trying to actually get this work on. No, in fact, I'm like 10 steps behind you. I'm just like, Chris thought this was way exciting, and I need to know what this is, and I have no friggin' idea. So that'd be well, I mean, it's I can kind of see the usefulness of it.

Like if you're in the middle of trying to do something and you're like, oh, I probably should have had this conversation in the custom model instead. You can bring it into the custom model if you forget to actually have the conversation in the custom model, sort of, it sounds like. Because when I look at this, I'm like, well, why wouldn't I just have the conversation in the FAQ generator versus tagging it into a separate conversation? Right. And so the idea being that you could reference that GPT as you were having a conversation midway through, you're like, hey, let's work with our ideal customer profile.

You maybe you you're working on it, and then you're like, oh, I need to talk to my co-CEO about this. So you could say, uh, you know, at Co COCEO, what do you think about whatever's going on? And it could invoke and bring that GPT into that conversation. So this has been a capability. It's been in an OpenAIS chat GPT for about a year now, uh, give or take.

And it's it's useful. It's useful to be clear. This is not, it's it's you know, it's it's not a novelty. But like I said, there are scoping issues with it that Claude in their implementation uh has, I think, done a really good job of addressing. And also uh, our our buddy Brian brings up a really good use case, which is I love bringing multiple custom GPT personas into one prompt and doing a virtual focus group.

That is a fantastic use case because you likely have custom GPT personas for maybe different segments of your audience or different members of your team internally. Bringing them all into that one conversation is actually a very smart way to use them. So thank you for that, Brian. Exactly. So Claude in their implementation, it's called skills.

Okay. And uh you will need to go into your settings, into your capabilities menu uh in Claude, and this can be on the web or in desktop, uh, and scroll down until you get to the skills section. If you have not enabled it, it will ask you if you want to enable it. Uh, it will then give you some feedback like hey, you need to turn this, this, this, and this on in order to enable it. And then after that, your skills become available, and you can reference them by using a hashtag.

So in chat GPT, you use an at sign, in Claude, you use a hashtag. And the reason you use the hashtag is because Claude reserves the at sign for agents, which is another show. I was gonna say, like, God forbid everything be standard across the platforms, but okay. Exactly. Um, and so what I've got uh uh I've got I've already built a few for you know, just for for testing when I was testing it.

One is to apply the trust and sets brand standards to any project. So hey, I'm in the middle of this thing, I made a cool infographic. I'm like, oh, it doesn't look like our brand standards. So I can invoke the skill and say use the trust insights brand skill, rebuild this thing, make it look the way it's supposed to, with the correct fonts and stuff like that. Uh I have a co-CEO one that I built based off of yours, Katie, but I added a whole bunch of stuff to it.

Um so now it is probably like uh it's got more frameworks and things available to it. Um I was gonna say, would real KD CEO approve of the things that you added that I haven't seen? Um conditionally, and of course the most important one, at least that that we recommend people have is an ask the ICP, um, which is actually loaded with seven different ICPs. And when you invoke it, it will say which one do you want to use because there's this is based on what was in our sales playbook. To make Claude skills, um, you just have a conversation with Claude.

So you there's there's no special interface, and they make it really, really simple. They in fact there is a a skill creator skill called skill creator to to allow you to build these things. Now, when it's done, you get a zip file. So if you ever used and installed a WordPress plugin and you get a zip file, you just drag it into WordPress and turn it on. Exactly the same thing.

And that's what I find makes this really, really so cool that you can't do with a GPT or a gem, is I could zip up co-CEO and just hand it to you, Katie, and now you have co-CEO in your instance of Claude. Uh if you had a separate one, and we all use the same one. I see what you's a weird thing. But I could put, say, my trust Insights brand uh thing. I could put that on Google Drive, and then if we had multiple instances of Claude, people could just drop it into theirs and they'd be up and running.

If I was really feeling enterprising, maybe I had a data analyst skill. I could package that up, put it on Gum Road and put it up for $49. And anyone who bought it would get the zip file, drop it into their Claude, and suddenly their Claude has that skill. You can't do that with GPTs and gems. So when they uh when you first of all, do you export it just from the window there, the normal interface?

No, you you export it from the builder when you build it. Uh-huh. When you re and then when it's passed off, is it visible? Can somebody open it up and see everything that's in there? Or does it remain opaque that they can't see what you actually build?

No, it's uh it's just a straight up zip file. Um, and like a WordPress plugin, you can open it up right up and just see what's inside the box. Uh I assume we're gonna be doing a demo of this. We will, but I'll show you what's inside the box. So inside the Ask the ICP, um, you get our IC, there's the ICPs and personas, which is straight out of our sales playbook, right?

So there's it's literally a copy paste from our sales playbook. So there's nothing, there should be no surprises there. And then Claude creates itself a skill system instructions uh that it uses so that when it is invoked, it knows what to do. Um, and so that's that's the content. Now the thing about the this really interesting is you can stuff a lot of stuff in here.

So it doesn't have to be just this. This is like the the the slim version of one, but you could put in multiple support documents, examples, you could put in templates, you could put all kinds of things in the same way you would for a WordPress plugin. And when and once Claude is told what to do with those things, it can it knows how to manage them and what and what to make of them. I mean, it strikes me as just another way to build a custom thing. You know, so I was building uh custom gems in Gemini earlier, and so you have the system instructions, but then you can also add additional documents for contextual knowledge.

And so this seems very much like that, you know, where you can add as much context as you need to, provided you know, you work within the limits. But like so, I built uh one that gathers business requirements. So you have the system instructions, and that is the contextual knowledge. It was the deep research I did on the SDLC, and I could also add in, you know, uh the way trust insights on boards or something like that. It seems very similar in that respect.

If I'm sort of breaking it down to its most simplistic thing, you have the skill, which is the system instructions, and then all of the background information that supports that skill for it to reference. Exactly right. So um, and that's that's how that's why this is it is is is valuable, right? Um, like I said, the thing I really like about I like the portability of it. The fact that I could put that thing on a SharePoint and sudden and add everyone at a company who's you know, we have one client that is a Claude shop.

Um, they can take their skills, put them, put them on their SharePoint, drop them in, and and they're up and running. So let's look at how to build one of these things. Uh the nice thing is um it's done right in the chat. There's no special interface, there's no uh crazy stuff, it's just literally just making things, but we still have to obey general best practices for software development because we're we're making a piece of software, so you can just wing it uh and say, hey, I want a skill that does this, but it's probably not going to be the best use of your time, and it's gonna uh there'll be a lot of of trial and error if you do it that way. So, Katie, is there something that before we get to that, Chris?

We do have a just a clarifying question uh from Jessica. So is this just Claude's version of a custom GPT or a gem? No, and that's where this gets confusing. Claude's equivalent of a GPT or Claude projects. I thought they were artifacts.

No artifact is what they call a canvas. Who's on first care? Exactly. Thank you, John. Um it actually gets even worse than that because OpenAI's project is different than a Claude project.

They do let's not bring them into this. No, let's not. They're they're we're booting them from the chat. We're staying focused on Claude. Okay.

Um, so we can sit if we can we're in this interface here, which is the standard stock interface, we can build any kind of skill. Um so what would you like to build? Um since we've been talking a lot about the SDLC for the past couple of days. So for those who don't know, the SDLC is the software development lifecycle. Actually, you know what?

I have a better idea because that could get unwieldy and we only have so much time on the live stream. Let's actually build uh a 5P skill. So that way if you're in a conversation and you're kind of going off the rails, or you're like, I forgot why I was doing this, you can invoke the five Ps to sort of like bring you back to where you started. Like, you know, here's what I know about your conversation so far. Here's your purpose, here's your people, so on so forth.

So I think that would be uh a useful skill. Okay. So let's let's do that. And here's how we're gonna do this. We're gonna go first to the Trust Insights website.

You can get and you can get this at TrustInsights.ai slash 5P framework. Uh we're gonna grab the 5P framework, the PDF version of it, and we're going to just drag and drop that right into our window here. So can we get it says upload fail due to a network connection? Great. Love that.

Okay, I'm gonna have to switch to the web version uh because it is unhappy with me. Um there should be no network problems now because you're literally on the internet. You jinxed it. In theory, I clearly have in theory, I have unless. Um, so while you're doing that, to answer Brian's question of can't we get these foundational model companies some branding help?

Ideally, yes, but they're named by developers and no shaded developers, but they have no business naming things. I can't challenge that, you know. It's almost like I've been there done that, John. You know, if your product name has a decimal point in more than four digits, uh uh back to the drawing board. So we're going the long way around, and I said, Oh, good, read the PDF.

Um, which is the least efficient way to do this possible. Um, but that's what the PDF says. So now we could say uh let's build this into a claude skill using and we're gonna invoke it, skill creator. Now, what we we want this to basically in conversation allow someone to say, have I thought like how how do you want it to work? Um I think there's a couple of different things.

So one is you know, you're in a conversation, you are starting a conversation. You're like, you know what? Let me use the five Ps to structure what I'm thinking. Or the other scenario could be I'm kind of all over the place. How do I bring this back?

Let's invoke the five Ps. Got it. Okay. The purpose of the skill is to help a user either frame out a thought using the 5P framework, or to organize their thoughts when they feel overwhelmed or distracted or all over the place and use the five Ps as a framework to reframe their thoughts. So this is a consulting skill, an advisory skill where the skill will be a trusted advisor to the user.

As always, we ask the the meta question. Ask me one question at a time until you have enough information to successfully complete this task. And let's see if we can get uh the skill the skill creator to invoke itself and to come up with some designs. All right, so you're gonna help me build this skill, reading the skill creator documentation. Its own skill creator is uh by the way, a very, very large uh thing of its own.

Okay, it says can you give me two or three examples of whatever user might say? Um those are really good examples uh that you've listed there. Like I'm feeling overwhelmed, or help me think this through. Yeah, like yeah, that's that's exactly what you would say. That's the idea.

Yay! Yay, it's pretty straightforward. Uh, how should Claude uh do this? Should it uh ask all of them up front, walk them through one at a time, assess which one that they need the most help with? Um, I'd say two and three.

So, first, they should assess which Ps the user needs help with, and then walk the user through each P one at a time. Okay. So, first assess which P the user needs help with most based on what they've shared, and then secondarily walk them through each P one at a time sequentially so that the skill has enough information to be helpful. When Claude is gathered enough information, what should the final output look like? A structured summary, strategic recommendations, an action plan, all of the above, something else.

I would say all of the above. So it should provide a summary, um, then the recommendations of you know, next steps, and you know, depending on what the thing is, there could be a lot of actions. So helping it prioritize like okay, you have chosen a platform, but you don't have a purpose, so start there. And what tone of voice should it have? Directive and prescriptive, Socratic and question based, balanced, empathetic and supportive, business and efficient.

Well, you know, Chris, with the gem that I built earlier today, you said it sounded incredibly optimistic. And I asked if I should make it sound more mean. So I mean I my gut says it should be somewhere between uh direct and prescriptive and empathetic. But if you want it to be all business and efficient, you know, I'm I'm fine with whatever. I don't get easily offended by machines.

Make it a good balance of directive and prescriptive and empathetic and supportive. That's not contradictory at all. Actually, well, I mean, you best you basically just describe me. All right. So now what it's gonna start doing is it's gonna start building the skill.

Skills are composed of a master skill document called skill.md, which is you can see in it's in the download that has uh some front matter, which is basically a like a little miniature table of contents and a summary, and then it has the system instructions that it builds. And then if there are supporting documents, it will typically include them, and then once it's done, it makes a little zip file. And then that zip file you download to your computer, and then it it would in whatever version of Cloud you're using, you can then drop that into your account and it becomes available across your account. Uh and that part's also really important is that no matter what version of Cloud you're using, it the skill becomes available there. It says, Why did you fail why did you fail?

Remove the template. Well, we're burner tokens here. Oh dear. It's a day ending in Y. It is.

It one of the things that's really interesting is that Quad in particular burns a lot of tokens. Like it it really it it overthinks everything. Okay, so it has packaged up the skill. Uh, and there it is, 5p framework. Uh, I don't know why it tries to show you the binary file because that's completely unhelpful.

Um makes you think that something broke. Exactly. Nothing nothing broke. Um, because if I look in my downloads folder, I have the 5p framework skill. Now, this is just the skill document itself.

There, there weren't there are no supporting files because the PDF uh was enough. It it put that in here. You can modify this before you upload it. You could modify it if you wanted to, if you forgot something. Uh so let's take that raw ver the the final version.

Let's go back into our settings here. Let's go into settings, capabilities, skills, and upload our new skill. We just drag and drop our zip file in. It says skill upload successfully, 5p framework. So we have uh that skill available.

Now, if I start a new chat, and let's say today I want to think through how to create a fun, mildly entertaining video game that's a weird mix of Donkey Kong and uh Contra. Now that's kind of a very strange uh uh request, and we'll see what it comes up with. But uh what I'm gonna prompt it then next is oh, I forgot. This is gonna be a uh an important project, so I probably should use a skill. Let's use the 5P framework skill.

And that we're gonna say uh available at 5P framework. I feel like that's a good plug for uh some governance around naming conventions. Make sure that you're naming things in a way that makes sense, and maybe perhaps, you know, uh depending on how large your organization is or you know how many different people have access to create skills, maybe create like a master dictionary or some sort of like inventory of the skills that have been created and what they're meant to do. Just a you know, quick plug for that. Um while that's running.

Uh quick question from LinkedIn user. I missed the start of this. Will be it a will it be available on watch again? The answer is yes. Go to trust insights.ai slash YouTube, go to the so what playlist, and all past episodes should be there.

So it did exactly what was requested, which is uh pulled up the framework and it looked at it and it started with it identified the the most important view. Why are you doing this? That is a great question to start with. What problem does this game solve? What gap does it fill?

Even entertainment has a purpose. This is about nostalgic arcade vibes, a specific gameplay itch, uh, or proving you can execute on a weird concept, right? So, what's driving this project for you? Proof that we can execute on a weird concept. It's giving Mario guns.

That's really what we're wanting but see. There's the empathetic part of it. It's like, no, I can see that you're trying to do something. So before we just shoot it down and say no, let's make sure that we've sort of really thought this through. Let's give you a chance to explain that stupid.

Well, I was thinking about making this for a customer. What if we use our ICP skill? Use ask the ICP to help with our people. Now let's see if we can invoke that second skill in the conversation. And this goes back to what uh our friend Brian was saying earlier is he likes to bring in custom GPTs into one conversation for a focus group.

This is a great way to think about why you would have different skills versus one custom project. Trying to have one thing that does it all is really never a great idea because it can certainly you can't do everything great every single time, but having individual uh purposes for each project and then bringing them into the conversation is actually really going to make whatever it is you're doing that much stronger. And it said, hey, I loaded your ICPs. Uh this makes no sense at all. Are we are you actually planning to make this Donkey Kong controversy for one of your customers?

Because that would be really interesting and weird in a good way. I mean, and this is about the time I would tell you to invoke the CEO to start asking, what the heck are you doing? Exactly. Maybe we need extra help from our co CEO skill to clarify. Let's ask them.

Use that skill. This is not gonna go well. This is gonna be funny if it just like straight up vetoes. It's just like, what the hell are you doing? Well, I mean, according to you, the CEO skill that you built was a lot meaner than me.

Yes, it is. Um, but you can see how to your point, Katie. If you have a mental palette of all the different skills that you've built, you don't have to use them all at once. You can bring them in as needed. And Claude also is smart enough to look at the conversation context and say, hey, I have this skill, like the ask the ICP.

So and by saying you're talking about ICPs a lot, maybe I should use the ask the ICB skill. Which is what happened the other day when I was using Claude, it brought in the co-CEO skill without my asking it to, which makes sense because I'm me and I was the one using it. Exactly. So the co-CEO says, I have questions. Um it really is me.

Is this a business project or is this a personal side project? Right? If it's a business project, what's the use case? Like, how does building a video game align with this at all? If it's a personal project, that's fine.

When you say for a customer, what do you mean? Is this is it for a client? Is it for sale, or is it people who play video games? What is this? So uh essentially, this is really is you.

It's basically saying I yeah, these are the exact questions I would be asking you. Exactly. It's probably and so these are the these are the Claude skills that this is essentially the what this is. You build these skills, then you you get the download, and the download again is nothing more than a zip file. So if you didn't want to use the Claude Skill Builder, all you would need to do is have a root file at the at the top of a folder called skill.md that has the expected structure, which by the way, this structure is identical to Claude Code sub agents.

So if you've used those, you know exactly how to build these. It's the the name, the description up top as a little bit of front matter, and then your system instructions after that. And then within there, you would have a docs folder with reference documents if you wanted, with templates, with whatever the thing is. Um for example, let's say you are a an agency that builds websites for customers, and you have a stock list of like 10 different templates that you use. You could have a skill that interviews the user and says, of these 10 different templates that I know that we produce for our clients on a regular basis, which template should we start with?

Have that conversation, and it can immediately bring stuff out of its own little archive here and say, Okay, we're ready to start working on template three or template four or what have you. If like there's a couple of assumptions here. Number one, you have to know that the skills exist. Number two, you have to know what the names of the skills are. So is there a scenario where you could say, you know, here's what I'm thinking: what skills are available to use?

And would it give you a list? That's a great question. Let's see if it knows. Because I don't know if it's aware of that. Uh Ted was asking in the chat, can we invoke skills within a project?

Yes, you can. And that's that's very, very useful. Oh, look, I can just ask it. I can it's Claude says I can make Word docs, I can make PDFs, I can make PowerPoint presentations. Shut the front door.

We're gonna test that one out. Let's test that. Okay, sorry, keep going. Uh I have the trusted sites business tools, the 5P framework, ask the ICP, the co-CEO, and the brand, and then I have some skills development stuff. So these are the things that I it says, that's what the what's available to you.

But that's really helpful because I feel like if you are it's hard if you're like midstream thinking, like in the middle of a conversation with your LLM, and you're like, oh, you know, I could really use some help on this. I think we had something that could do that. It's nice to know you can ask, is there a skill that can help me with this? And it might it might just give you this list again, which is fine, and then you would have to invoke it. Or, you know, theoretically be like, yeah, you have an ICP.

I'm gonna bring it right into the chat. My other question, and you may not know the answer to this, is I'm assuming the answer is no. Like, let's say I brought my ICP uh ask the ICP skill into the conversation and then decided, you know what? It's too constraining. I don't actually want the ICP.

Can you boot the skill from the conversation? No, because at that point it's been loaded into the context window. Okay, that's what I thought you were gonna say, but I figured it's always worth uh asking. Yep. Let's uh I'm now curious about the PowerPoint thing.

Let's say Yeah, I am too. If I provide you with a slide deck template, can you use the PPTX skill to modify it? Uh let's let's ask, and I'm gonna invoke it and see what it says. I mean, I can't see our audience, but raise your hand if you've been waiting for this one. It says upload your template, I'll read the skill, I'll modify it, and it'll deliver the results.

All right. Well, I have our Trust Insights master slide template here. Um, and let's see. Um here is our slide template. We're gonna turn this into a 10-slide sales presentation.

And our uh target customer that we're gonna be giving this presentation to it we'll call uh skill point consulting. And we are pitching them on an AI project. Let's go ahead and get started framing out what this presentation might have in it using the PowerPoint skill. Let's see if it works. Let's see.

Read that, five up the press factors, uh uh examine your template. It's funny they should have the token wheel like spinning berserkly. It does have a usage meter. Uh it's it's in the preferences. You can see how much of your weekly allotment you've burned up.

Fuel burning. Mm-hmm. Let's see, it's creating a thumbnail grid of template slides for visual analysis. Now let's see the visual thumbnails. I think we're all holding our breath hoping that this actually works for once.

Yeah, for folks who don't know, by the way, this is the document that is in question. It's just literally our blank uh template we got from our designer. Let's see. Section two, design elements. So architecting presentation structure.

Show me your thinking. I want to see what you think here. Plain text, 10 slides. Today's data is correct. The challenge.

Section divider. Okay, so it knows that our section dividers exist. Which is interesting. Well, the if you look at the PowerPoint, that's the slide template that's selected. Yeah.

Because each slide is, you know, you choose a template from PowerPoint. Like, I want, you know, uh a header slide or text and body, or you know, two side by side or text and image. So it's probably looking at whatever that background information is slide by slide, uh, and saying what's available to me to use. So I'm assuming, and this is a big assumption, that unless you give it all the different slide templates that you may want to use, it's only going to use what you give it. So if you want to use um section headers or dividers or whatever they're called, but you don't give that to this, it's not necessarily going to say, oh, and by the way.

That's a big assumption. That is a big assumption. Interesting. So it's got a combination of Python code on the back end and object manipulation to try to essentially frame out. I see what it's doing.

That's very clever. So what it did is it took apart our slides, and because PowerPoint a PPTX slide is nothing more than a big X uh XML document with some images attached to it. So it basically deconstructed that, figured out what fields are where on our slides, and was able to understand the general structure of our slides and what goes where. Because every I every object in a PowerPoint presentation has an ID number. So like the text field is ID12, um, uh an outline is ID 14, etc.

And so it's essentially disassembled that, looked at our content, put together an outline in a JSON object, and now it's going to it looks like it's trying to reassemble the slides with the changes to the XML tree. Yep, there in fact, there that's literally what it's doing to try and rebuild the slides with the changes to them. I don't know how long this is going to take, but that's a very clever way of doing it. I've all the only other service I've seen that does something like this is Moonshot AI's uh Kimi K2 model, which you can't upload a template, it has its own templates, but it it knows what regions of those templates it's allowed to modify. This is another uh sort of plug for this is great, but first you still need to go through probably the 5P framework to figure out what it is you're trying to create.

You need to have the template, so you need to have all that background information first. Like here's my template, here's what I'm trying to accomplish, here's what the purpose of this slide deck, here's who it's for, you know, here's how we're gonna present it. You know, we want to have you know slides that include slots for video, or we're gonna cut out of the slide deck to do a live demo, and then the overall so like you still want to do your upfront requirements gathering before you start using a skill like a slide builder, just like you would, you know, any other PowerPoint deck. Um, when I was building our AI Ready Strategist course, I had that outline, like slide by slide outline, and then I just had to fill it in. And Chris was able to create an application that did that for me automatically and got me, you know, 60% of the way there, which was super helpful.

But I would still, you know, for this, I would still be like, here's our slide deck template, here's my you know, 30 page outline. Now can you put it together? And that would definitely be helpful. Um, because the widget, Chris, the application that you had created didn't take the full template. It was just a bunch of bank blank slides, which was still really helpful.

It says the presentation is complete, and now it's got to move files around. Sure. Um and see it's providing a an outline of the presentation itself. Wild is it'll be interesting to see what the output is because if it's just an outline, we didn't need to use the skill for that. Right.

Um it will be very interesting to see that. And I will be honestly surprised if it finishes it all. Um because knowing the back end of what it's doing, this is a very, very, very computationally intensive process. I mean, this is probably churning up you know a kilowatt of power just to just to do this. So let's add a little bit of context to what that is that you're saying.

So anyone who's using any of these large language models, uh, you know, we've talked about how each of these large language models use tokens. So every word, every ask, everything you're trying to do with it. Let's say you have a bag of tokens that renews every week. As you're asking these large language models to do something, you're withdrawing those tokens from whatever your allotment is. And I think that that changes with whatever it is you're paying.

So the paid, so for the free, you probably have certain limits for paid, so on and so forth. And so it sounds like Chris, if I'm following, asking it to do a skill like create a PowerPoint is very when you say computational heavy, it means if you start with a thousand tokens, you might end up using 999 of those tokens just by doing this one thing. Yep. So in our cloud account, we are on Claude Max, uh, which is the most expensive plan. Oh, good.

But before um, before we started, we were using about we were on 12% of usage for the week. Because this is the same model we use for a lot of coding. So just the 10 slide deck alone used 2% of the max plan's computation, which is not an insignificant amount of compute that it used. However, it successfully figured out our slide templates, put it in the correct fonts, use the section headers appropriately, and came up with essentially a decent starter that's in our template. Yeah, that's not terrible.

Yeah. It definitely could use some more images and graphics and things like that. And you know, it violates the the general standard of no more than nine words on a slide, but it's a start. I'm thrilled it opens. That's really I was like, that's not gonna open.

Like you can't mess, you know, you can't plug the metadata like that and make it work. So I usually well, and I feel like everyone has different standards for how they create slides, so that's probably something additional you'd want to give. Like, here's our brand guidelines, or here's my best practices for creating a slide deck. Because Chris, you and I do it very differently, and so we'd want to include that kind of information in the prompt up front. Like, you know, you must not uh use more than nine words on any given slide, or you know, add clip art wherever applicable, which is your favorite thing to do.

Exactly. Um but again it it managed uh 10 slides. It did at the very end of the process, it did crash. Um, so uh it it skated through just barely to the end before it it blew up. Um what didn't it do?

No, it did everything, but the it then gave me a lost connection with Claude uh error message. Um so I don't know that I would try a 200 slide epic in here. I think that could go very badly. Well, and that's sort of what I was uh the point that I was getting to was so when you're saying that tasks or skills like this are computational heavy, and for us at Trust Insise it used two percent. Imagine you know, exponentially more at an enterprise-sized company where everybody's trying to get their weekly reports done or their monthly reports done, and all of those slide decks are you know in the hundreds of slides, and everybody's using the one same account all at the same time, like that's when you're gonna start to see those tokens get eaten up really quickly.

Exactly. Um the difference between the the plan we're on and your average $20 a month plan is 20x. So if we use 2% of our compute uh budget for the week, because Claude uh meters weekly, uh, on the lowest paid plan, um, you will have you would have used 40% of your week's compute on a single slide presentation. And that is significant. That is significant.

That is that is problematic if if you were also using it for other things because everything counts against your quotas. Uh and this is true of all the systems and all the vendors, but in general, these skills are super handy, and you can see how you can invoke them. And if you think about it, like those basic document skills, if you're having a long chat and conversation, you could at the very end say, like, hey, turn this into a slide deck for me. Right. So not just create a summary, but actually turn this into a slide deck.

Or you're working on uh some some accounting to say put this into an Excel spreadsheet for me. Um, and it's capable of doing that in the same way that in Gemini, you can say put this into a Google Sheet for me, and it will also do that in there. The skills are portable, which is super handy. Uh, you can build build once and hand it to as many people as you want. Uh I have already seen people on LinkedIn offering their their Claude Ultimate Skills pack for $99.

Um, which, as we saw, is just nothing more than their previously prompted their previous packs of prompts they used to sell. Um but it it is it is super useful, and we didn't cover it because uh I would have to go inside it up, but it the skills are usable within Claude Code. So if you're using the Claude coding tool to do batch tasks, uh I'm gonna be testing this myself. I have Katie's you know writing style, and I have several hundred transcripts of YouTube minis that I've done over the years. I'm gonna put 10 of them through and see if I can have it rewrite my transcripts as though Katie was writing about those particular topics and see what comes out using Claude code and a write like Katie skill.

Hey friends, here's an agentic AI. Because for some reason you think that's how I sound. I don't think that that. That's how the machines think you sound. Well, we'll be interested to see the results of that, and perhaps we'll post uh some of those results in our free Slack community, uh, which you can get to at trustinsights.ai slash analytics for marketers.

Free to join. Always a good time. Umhinged questions every day. Yeah, different questions. Fridays are unhinged.

Tomorrow should be interesting. Umhing. No, this is true, but on Friday we purposely have unhinged questions. Any final thoughts on Claude skills before you go off and start creating your own. Um, I don't know.

John, what kind of skills do you think would be useful from your perspective uh on the sales side on the biz dev side? Well, the the thing is that that that's the first time I've actually seen a slide builder that I was like, okay, yeah, that could be useful. Like even I don't even care if it burns up all the tokens on a side account. If I can throw an outline in there and get 50 slides with the outline plugged into it, like that's that's huge. So uh I'm I'm taking the win there.

All right. Well, that's gonna do it for this week's show, folks. Thanks for tuning in, and we will catch you all on the next one. Thanks for watching today. Be sure to subscribe to our show wherever you're watching it.

For more resources and to learn more, check out the Trust Insights podcast at Trustinsights.ai slash TI podcast. At our weekly email newsletter at TrustInsights.ai slash newsletter. Got questions about what you saw in today's episode? Join our free analytics for marketers Slack group at TrustInsights.ai slash analytics for marketers. See you next time.