Everyday AI Podcast – An AI and ChatGPT Podcast - Ep 736: ChatGPT Skills: How to Use The New Feature from OpenAI and The Best Use Cases

Episode Date: March 18, 2026

Imagine your work life without templates, SOPs or copy and paste. 😱Maybe that's how you've been working in ChatGPT, too. Yes, ChatGPT's custom GPTs and Projects are useful, but left... something to be desired. The solution? ChatGPT's newest release: Skills. Join us as we put AI to Work on Wednesdays and show you the ins and outs of Skills and the best use cases and our secrets to get started. Newsletter: Sign up for our free daily newsletterMore on this Episode: Episode PageJoin the discussion on LinkedIn: Thoughts on this? Join the convo on LinkedIn and connect with other AI leaders.Upcoming Episodes: Check out the upcoming Everyday AI Livestream lineupWebsite: YourEverydayAI.comEmail The Show: info@youreverydayai.comConnect with Jordan on LinkedInTopics Covered in This Episode:ChatGPT Skills Feature Introduction & OverviewReusable Workflows for Business ProceduresChatGPT Skills vs. Custom GPTs & ProjectsTeam Access: Enterprise, Business, EDU PlansAnthropic vs. OpenAI Skills CompatibilityHow to Build and Deploy ChatGPT SkillsSkill Sharing and Collaboration for TeamsBest Use Cases for ChatGPT SkillsPractical Setup and Implementation WalkthroughEditing, Importing, and Managing ChatGPT SkillsTimestamps:00:00 "Building ChatGPT Skills Live"06:05 "Skills: Reusable Workflow Automation"09:52 "Reusable AI Skills Explained"10:59 Anthropic's Pioneering AI Skills15:51 "Skills Access on Team Plans"17:51 ChatGPT Skill Creation Tips21:29 "Maximizing AI Tool Utilization"25:18 "Skill Setup and Integration"28:25 "AI Skill Editing & Updates"32:02 "Streamline Workflow with AI Skills"35:59 Streamlining Operations with GPT Skills36:55 "AI Updates and Features"Keywords: ChatGPT skills, reusable workflows, AI SOPs,Send Everyday AI and Jordan a text message. (We can't reply back unless you leave contact info)

Transcript
Discussion (0)
Starting point is 00:00:00 This is the Everyday AI Show, the everyday podcast where we simplify AI and bring its power to your fingertips. Listen daily for practical advice to boost your career, business, and everyday life. Meet Firefly AI Assistant, now live in Adobe Firefly, the all-in-one creative AI studio. Just describe what you want to create and the assistant handles the rest, orchestrating multi-step workflows across Photoshop, Premiere Express, and more in one conversational interface. You direct the outcome. The assistant accelerates execution. Imagine your day-to-day business life if you didn't have things like templates,
Starting point is 00:00:51 SOPs, or even copy and paste. Well, maybe you've been using chat GBT without the equivalent of those things. Yes, we've had for a long time inside of chat GBT, very helpful features like custom GPs and projects. But I still think that there was a certain level of modularity and scalability that we were maybe missing. And Open AI may have just changed that with their recently new feature that they released, which is ChatGPT skills. So on today's show, that's exactly what we're going to be diving into to show you how to use them and some of the best use cases. So here's the big picture.
Starting point is 00:01:39 Chat Chb-D skills are essentially reusable workflows that teach chat chbti how to follow very specific business procedures consistently that you upload and you can do them in natural language. So they essentially package repeatable tasks, your company's domain knowledge, and executable scripts into a more modular format. So in the same way, like you might have very strict and refined and defined and defined templates and SOPs at your company, the same thing is now available with chat GPD skills. And if you're hearing this skills thing and you're like, wait, hasn't this been around for a long time? Yes, technically anthropic created and popularized skills back in the fourth
Starting point is 00:02:29 quarter of 2025. And now, Open AI has finally adopted them inside of chat GBT. So previously, skills were available on the developer side if you were using them in the API or inside codex like I was, but now they are available inside of chat chib-t. Downside is even if you're on a paid $20 a month plus plan or a free plan or the $200 a month pro plan, you're not going to find skills there. So there's the caveat. Right now, these are being rolled out and maybe even positioned as teamwork skills, right? because they're only available if you are on the business enterprise team or EDU plans.
Starting point is 00:03:14 So on today's show, here's exactly what you're going to learn if you stick around for the next 20-some minutes. You're going to learn exactly how to build and deploy your first chat GPT skill live because it is put an AI to work on Wednesdays, our weekly segment where we go hands on and we're going to be doing this live. Then you're going to be able to explain the differences between regular custom GPTs, and these new skill workflows. And at the end, I'm going to share with you some example in, I think, really good use cases of skills. And along the way, I'm actually going to give you what I think is the one cheat code, the best way to actually start using skills today and not really just following the leader
Starting point is 00:03:57 and doing what everyone else is doing. Does that sound good? You're ready to maybe start implementing, you know, templates and SOPs and essentially copy and paste? into you or your company's workflow. It's exactly what we're going to be unwrapping today. And apparently you all wanted this, right? So if you aren't listening on the podcast and you're like, man, this guy is always covering
Starting point is 00:04:19 these, you know, random things. Well, I asked you what you wanted to hear, right? So on Wednesdays, like I said, we always go hands on doing some live demos on a newer update model release feature. And, you know, I put it up for a vote. And overwhelmingly, our audience said, let's go and learn skills. All right, well, let's learn together. And if you're new here, welcome to Everyday AI.
Starting point is 00:04:42 My name's Jordan Wilson. And if you don't know any of this, we do this every single day, not just Wednesdays. It's a daily, unedited, unscripted, live stream podcasts and free daily newsletter, helping everyday business leaders like you and me keep up with the nonstop breakneck developments of AI. I tell you what's noteworthy, what matters, and you take that information as your cheat code to grow your company and career. So it starts here, but make sure you go to our website at your.
Starting point is 00:05:08 your EverydayAI.com. We're to be recapping the highlights from today's show, as well as giving you all of the other AI news that you need to know to stay ahead. All right. So chat GPT skills. I think this is a big shift into maybe how teams are ultimately using Chad GBT. I kind of think of it as the, you know,
Starting point is 00:05:29 kind of a line in the sand. There's the old chat GBT that was more of a friendly assistant that it's a chat bot. Now, today's version of chat GBT, it's a full AI operating system, right? Because we've gone from just simple prompts to then it got a little bit more scalable, right? And you didn't just, you know, with prompts, it's essentially one-time instructions that live and die there. And then we over time got things like memory and chat history, which helped. And then we got custom GBT's, you know, now almost two years ago, which allowed you to create a custom version of chat GBT.
Starting point is 00:06:03 And then we got projects, right, a way to better org. organize your chats, but also share files and instructions across those chats. And now I think skills are a bigger step in that same direction. So I do think skills kind of build off of some of the momentum that custom GVTs and projects and memory helped create. And I really do think it's the one thing that encapsulates all of those things together, which will unpack a little bit more. But this is essentially skills are reusable workflows that your company can package
Starting point is 00:06:35 together, you know, your SOPs, templates ready to go. So a skill essentially just tells Chat ChbT how to execute a highly specific workflow the same over and over without you, the human, needing to re-explain and course correct each and every time. So here's what's new. So like I said, they launched right now in beta just for teams. So the enterprise plan, business plan, and EDU workspace plans. they are powered and made possible by the new GPD 54 model.
Starting point is 00:07:09 That's the technical backbone. And eventually we've been told that you'll be able to convert those custom GPTs into skills. It's not available yet, but it does seem like that's going to be rolling out to enterprise plans. But it's very easy to create these, you know, from existing GPTs projects because it's super simple. And I'm going to show you how to do all of that. And the cool thing is, like I talked about, skills really build off of everything else that you already have because skills can use apps. You can use skills and GPTs at the same time. And skills essentially cover the basic functionality and then some that you would get inside of projects.
Starting point is 00:07:51 So now that you know how skills are kind of implemented in chat GBT, let's maybe hit rewind and be like, what the heck are skills? So skills, they're just reusable workflows. Think of it as SOPs, your company's standardized playbook that teach AI agents exactly what they need to do each and every time a user invokes those skills. From a visual standpoint, they're folder bundles, right? Each skill is a folder with a, you know, skills.markdown file containing core instructions. So you might look at skills and be like, oh, that seems technical, not for me, but you don't have to write them by. hand, right? Chat GBTbt will write your skills for you conversationally. And there is something called progressive disclosure. So at first, the AI just reads the skills name and descriptions,
Starting point is 00:08:42 and then it fully loads those all the instructions and the executable actions and files only when this skill is triggered. So from a visual perspective, think of it like this. And this will maybe help even our podcast audience better understand, right? I am going to be sharing some visuals. And yes, on Wednesdays, maybe those are the shows that you should go home or at the office. Go to our website, Your EverydayAI.com. We always have the video version up. You can watch that for free.
Starting point is 00:09:12 But I'm going to do my best to describe visually what this looks like. But skills in the same way, if you've ever downloaded a program on your computer, right, and you unzip it, you unpack it. and there's usually multiple files in there. And those files essentially tell the computer how to run a program. And there's multiple different files, right? So that's essentially what a skill is. It is a kind of a parent folder with, you know, different markdown files, different references, assets.
Starting point is 00:09:41 So these are all things that you can customize in natural language. ChatGPT can do it for you. You can upload files, do all of these different things. So literally think of it, you probably have a similar setup like this, maybe in your organization, where it has all the different templates. Maybe there's an instructions file that says how you use the templates, you know, brand guidelines, all of those things. Those are the exact, you know, kind of the exact backbone that skills kind of can do
Starting point is 00:10:07 for the front end AI. So essentially, it uses the agent skills open standard. So I'm going to get into how these are reusable, both bringing different skills inside of chat, EBT skills, and how you can ultimately do the same thing and use them elsewhere. But how you use them, well, Chad GPD will automatically recognize relevant skills based on their name, or you can also explicitly activate a desired workflow by mentioning its name. Right. So it's not the same as a GPT where you go and find the GPT or if you're in a normal chat, you can click the at button and start typing and then find your GBT and click it, right? It will sometimes know or infer that you
Starting point is 00:10:53 want to use a skill or you can call it out by name and use the skill that way. So like I said, Anthropic made this happen, right? And you're probably, if maybe you're wondering, well, what's the difference between anthropic skills and chat chbtee skills? Well, the good thing is they can be used interchangeably, right, which is a huge benefit for people that are maybe using both in their day-to-day work or maybe you're trying to transition from one to the other. But essentially, Anthropic created and popularized this skills format.
Starting point is 00:11:22 and now it's widely used across the AI landscape. And Open AI essentially just adopted the same open standard after Claude's success. So some of the differences, you know, skills have only been out in Chachapit for a week and a half. And they've been out for, you know, now like five and a half, almost six months, I believe, inside Anthropic. So when Anthropic rolled out skills, they were a little rusty, right? But that's what happens when you are literally blazing the trail. Right now, anthropic skills are much. But I will say, open the eye skills to start with really, really good.
Starting point is 00:11:56 And for me, I actually prefer the UI, the user interface and user experience of creating and editing skills inside of chat chbtee versus claw. It's to me a little easier, even though, you know, I do think anthropic skills implementation is a little bit more mature and maybe a little bit better in terms of very difficult workflows right now. I do think chatubTs are a little bit easier to use and easier to monitor. modify and just simpler to understand, period. So now let's talk about the differences between a skill versus a custom GPT or a project.
Starting point is 00:12:35 So custom GPTs are great. Don't give me wrong. But they lack a lot of the depth and breadth that you get in a skill. Well, mainly because you can use a skill while you can use a skill while you. you're still using a GPT. But think of it like this way. I like to think of skills as describing an entire task in its entirety, right, where I think custom GPTs maybe help get you closer to a goal, but they're not necessarily at least
Starting point is 00:13:07 by default, really meant to chain all of these agentic skill sets together one after another, where that's exactly what skills are, right? Like a very, think of the most detailed, you know, SOP that you've seen at your company. think of skills that way, right? Where it can interact with other apps. That's the great thing. Skills can interact with different apps. You can use them at the same time as you're using GPTs,
Starting point is 00:13:31 and then you do get the same features and functionalities that you get within projects. So by default, I do think skills are instantly better, more useful, and hold more utility than GPTs and projects. Although there is a lot of crossover, I think ultimately skills might be, the kind of feature or function that takes off the fastest and goes the furthest. And you can also use multiple skills to combine automatically to complete a complex and multi-step project more
Starting point is 00:14:02 effectively. All right, let's get in and get our hands dirty. We're going live, but first, I got to take a drink. Quick word from our partners. Here's a harsh truth. Your company is probably spending thousands or millions of dollars on AI tools that are being massively under utilized. Half of companies have AI tools, but only 12% use them for business value. Most employees are still just using AI to summarize meeting notes. If you're the one responsible for AI adoption at your company, you need Section. Section is a platform that helps you manage AI transformation across your entire organization. It coaches employees on real use cases, tracks who's using AI for business impact, and shows you exactly where AI is and isn't creating value.
Starting point is 00:14:51 The result? You go from rolling out tools to driving measurable AI value. Your employees move from meeting summaries to solving actual business problems. And you can prove the ROI. Stop guessing if your AI investment is working. Check out section at sectionaI.com. That's s E-C-T-I-O-N-A-I.com. All right. Let's get our hands dirty, shall we? All right. So here is the part of putting AI to work on Wednesdays where sometimes things can go wrong. So we'll see if they do or not. All right. Ready? Here's my little secret on what you should be doing. So I'm going to get this started and then we're going to come back to it as we go.
Starting point is 00:15:39 And I'll tell you exactly what's going on. All right. So I just hit enter because, yeah, I like to do these things live. Sometimes they work well. Sometimes they don't. So if you want to find skills, remember you do have to be on a, business plan. So business, enterprise, edu, etc. Essentially a team plan. This is not available yet for individuals. I hope they do make it. So in the bottom left hand corner, you're going to
Starting point is 00:16:04 click your icon and then you should see a skills section, which kind of looks like the icon is kind of like a little brick. So from there, I've been cleaning up my skills, you know, had a lot of them that I was demoing around. But then you're going to have a new skills section, which might look and function a little bit similarly as the GPT section. So you're going to have ones that are installed. So for you, by default, you're only going to have the skill creator, which is Open AI's skill that, yeah, it's kind of meta. It's a skill that helps you create skills, all right?
Starting point is 00:16:37 And like I said, there's skill repositories, right? You can download skills, all those things, you know, in other places and then upload them. So you have your installed skills. one's created by you and then skills that you can share across your team. That's the big unlock here, right? I'm just showing off an individual use case here, but imagine sharing these across your team in the same way that you can share, you know, GPTs, sharing skills.
Starting point is 00:17:05 You know, again, think of it. SOPs, templates, copy and paste, time savings, sharing that across your organization is the big unlock. And then in the upper right-hand corner, you can click new skills. So you can create a skill with a chat, So you can do that conversationally. You can do it even without using the skill creator skill,
Starting point is 00:17:26 although that would be used inside of a chat. I know that's confusing. So there's technically four different ways. You can do a chat naturally without using the skill creator. You can do a skill using the skill creator. That's number two. You can just create with the editor from scratch. So not in the chat interface, a little more technical, right?
Starting point is 00:17:44 And then you can edit them as well. Or you can upload them from your computer. All right. So that's the four different ways that you can create or import skills. Now, let me take you to the little cheat code here because I think the natural human tendency is to go find the most popular skill and use it for yourself. Not always a good idea. Don't always go out looking. You know, don't be a solution looking for a problem. Instead, use something very powerful inside of Chachibati, and that's its ability.
Starting point is 00:18:19 to tap into its memory, how you use chat chabit, your chat history. And then I would ask it some questions. So I'm going to show you exactly how I'm going to, well, reset this up. I already had a setup. And I deleted it so, you know, could not show too many things on my screen here. But I essentially have two working files that I use across all different large language models. There are two markdown files, which is essentially just a text file, right? It's about everyday AI and then about me, my preferences, how I work, all of that good stuff.
Starting point is 00:18:53 My brand voice, all of that is in these two files. I keep them updated constantly. So then I can at any point, you know, use them. They're in my Google Drive so I can connect dynamically, et cetera. So I said, based on the attached files, please carefully go through my day-to-day workflow, understand what chat chip-tip-t skills is capable of, and briefly suggest the 10 most helpful skills I should implement and how they'll save me time. You could also just say, hey, based on, you know, your memories about me and my chat history, same thing, because chat GPT, AI systems
Starting point is 00:19:26 are so good at knowing things about your own personal work style that you might not even know about yourself. So I say when you're trying to figure out what skills to start with, see where you're going wrong or what you're constantly using chat GDPD for or what your ultimate, you know, deliverable or artifact is because then you can kind of reverse the process and start with what you actually need, not just what others are doing. All right. So I'm going down here. We're going to pick one. So there's the daily AI, right? I'm not going to read all 10, but it gave me the daily AI research brief builder, the Jordan hot take extractor. Excuse me, got a cough. Oh, fantastic. I've been six since November, more or less.
Starting point is 00:20:13 If I sound nasly, yeah, I've just been sick for months. All right. Jordan Hot Take Extractor, I kind of like that one. All right, there's a lot of other ones. I'm just going to use this one, and I'm going to say, let's do it number two. All right, so give me 10 suggestions. All I have to do is say, let's build number two.
Starting point is 00:20:31 It's probably going to go through and come up with some clarifying questions, which is good. All right. So it's asking input. What will I usually give it? I'm going to say podcast transcript. All right. Then I'm going to say the output. What should the output produce every time?
Starting point is 00:20:49 So I'm going to say it's saying five to 10 Jordan style hot takes, three episode angle titles, five. I'm just going to say all of those. All right. And then three, where should I pull the context from? Should this skill work only from the pasted uploaded text or also search you're connected? So I'm going to say only from the pasted. uploaded text. All right. So simple enough here. I'm using the skill creator. And again,
Starting point is 00:21:20 you can look at the bottom in the prompt input box. It's highlighted the skill creator. But even if you just say, help me turn this into a skill. So, you know, one thing aside from going through this process that I'm showing you right now, another great example is go look at a recent chat that you had with chat chabitie there was a lot of back and forth but you finally got something just the way you wanted hopefully you went through the prime prompt polish and refined cue uh right because then you got a much better output uh so all of our people in our inner circle probably went through and did that oh excuse me again there we go uh yeah keep muting my mic so much it's it's unedited unscripted y'all this is real all right so go through that chat that you went through and you finally got a
Starting point is 00:22:15 great output and then just reverse engineer it and say i want to turn this into a skill outline every single correction mistake pivot everything we went through and turn this into a skill right so at any point you can just click the backslash button and then start typing skills and then go to the skill creator. All right. So this is probably going to take a couple of minutes to build. All right. I'm not just going to keep blabbing on.
Starting point is 00:22:45 So again, instead, I'm going to pivot here a little bit and go into my Claude account. So the great thing, this uses the agent skills open standard. So what that means is, well, skills kind of work anywhere. Right. So, you know, you got to tip your cap to Anthropic between, you know, you know, So MCP, the model context protocol, agent skills, a lot of things that they've kind of innovated and have helped make the AI community at large better. So I now am in my Claude account.
Starting point is 00:23:18 So if you want to know where this is, they just moved it kind of recently. So it's on the left hand side. You go to customize skills, and then you can choose your different skills there. So in the upper right hand corner, I selected AI researcher. all right, and I'm going to download it. So you can click the three little dots in the upper right hand corner, go into download. So now I'm going to go back into Chad Chb-T in the skills section. Okay.
Starting point is 00:23:45 And I'm just going to click new skill. I'm going to say upload from my computer. And then I'm going to go ahead and upload that one from Claude that I just did. And within a second or two, it says skill uploaded. Pretty, pretty simple, right? So now I'm going to go ahead. There it is. AI searcher.
Starting point is 00:24:06 So at any point, you can click on a skill, even the Open AI skill. And you can see exactly how it's set up. You can read it. So for the skill creator as an example, here's the skill markdown file. Here's the references folder with the different files that are in there, the scripts, the different Python scripts, the license text. So like I said, in the same way, you know, if you do any coding, right? Obviously, you probably understand this. But even if you know, if you've ever downloaded a program to your computer and there's all these different files and folders and license texts, right? It's kind of like that.
Starting point is 00:24:41 You are literally taking and bundling up your company's knowledge, your files, you know, different things, different access points and putting them all into a essentially an executable skill folder that then open AI, Anthropic and others can use. All right. So on this skill section, there is a nice little shortcut. You can hover over any of these skills, and then you get the option to edit them. But you can also click the little chat icon, and then I'll just start a new skill right away. So I'm not even going to do anything here. I'm just going to say, give me 10. I'm going to say give me the top 10.
Starting point is 00:25:19 All right. That's all I'm saying. So again, not having to go through, you know, refine all of those things. It's going to go through. All right. So when I was trying this earlier, it actually did a little bit better job. I think it was because I was in the extended thinking. I'm going to try this again.
Starting point is 00:25:40 All right. I'm going to say, all right, load my skill up there. And I'm just going to say top 10 stories. We'll see if it comes back with clarifying questions. There we go. So now it's using the skill. So I don't have to go in and tell it the different things that I want. So, you know, let's just take a quick look at this skill that I have.
Starting point is 00:26:07 So this skill, I can go in and edit it. So the AI searcher skill, okay, interesting. So yeah, sometimes things don't work well. So even though the skill is working, I can't go in and edit it for some reason. That's interesting. So I'm going to go into the skill creator. Or now I'm going to go into my weekly triage and edit it. All right.
Starting point is 00:26:31 So this is a different one that I created. So in this instance, I have it. The description is it triages the user's upcoming week by reading Gmail and Google Calendar, identifying important meetings and commitment-driven, driving emails, doing lightweight internet research to fill context gaps, and producing a one-page weekly plan, plus a scored backlog and recommended time blocks. So in this instance, I can go through and just change these things very simply, Right. It's a Word doc. I can go in if something's not working, I can just edit it live from here. Right. And the good thing, like I said, with skills, they can access files and folders, right?
Starting point is 00:27:13 Over here in the files, I can upload different files and then have it reference those things. I can go and change anything that I want to right there in the skill editor. So it's very easy to manage and edit these skills. All right, there we go. And just like that, it gave me my 10 stories. So it went through the AI researcher skill, which if I look at the Claude skill here, right, it searches for and presents the latest AI news from the past 24 hours, focusing on large language model updates, Gen.
Starting point is 00:27:48 AI breakthroughs, developments from AI companies like OpenAI, Anthropic, Google, Meta, Microsoft, Apple, NVIDIA, Amazon, Salesforce, Adobe, IBM, etc. right? So I didn't have to go through and describe, right? It's a lot more than that. I didn't have to go through and describe all these things, what I wanted. I just invoked that skill and I said top 10 stories. And it knew all those things because I had technically be a clawed. I had already taught it. And I had gone through and made corrections and made updates and said, here's how we want the format. I want the format very simple, very easy with a heading, with a date, what happened, why it matters. That's it. Right. So it went through and it did all that for me. And this new AI searcher skill went through, worked correctly, right? I'm looking through, yeah, it got everything. It got all of the most recent stories from the last day. So did a great job. All right.
Starting point is 00:28:42 Now let's see if my skill is done. So to go back and remind you before we got slightly sidetracked, that's what AI work on Wednesdays. Yeah, we just do things live. We bounce around. We try to explain what's going on. All right. So remember, I uploaded, and I recommend you do this, I uploaded these files about me or just say based on my chat history, suggests 10 different things. The one that I liked was kind of the Jordan Hot Takes extractor. Say the truth, I actually use this a lot in Claude, and I'll probably start using it a lot in chat chivity.
Starting point is 00:29:15 A lot of times I'll have, you know, longer podcast episodes and I'm like, wait, I thought I had a pretty good take in there. But I completely forgot what I said sometimes, right? this thing's unedited, unscripted. Yeah, sometimes I have notes to help keep me on track a little bit. But I'm like, man, I had a great point about ABC. And now someone's asking me, how do I find this? Right. So I can run this automatically and just put them all into a file or folder.
Starting point is 00:29:39 And then I can more easily find them. So it went through. It created this skill, right? I gave it some feedback. And then I go to the bottom. And there we go. It says the Jordan hot take extractor. Right.
Starting point is 00:29:52 So right now it's a draft. I can click on it and then I can click install. So very easy, one click install. Like I said, you don't have to know anything about Markdown. You don't have to know anything about YAML, about Python, nothing. If you can go in and talk to chat, GPT, and approve some things and iterate, have a conversation. If you can be creative and strategic, you can start using skills in minutes.
Starting point is 00:30:17 All right. Then I'm going to go ahead and try in chat. So here's what I'm going to do. I'm not even going to give it any instructions, right? That's the whole point. I'm going to go in on my podcast episode here from the other day. I'm just going to copy and paste this entire transcript. Let's see how it works here.
Starting point is 00:30:35 Let's see what model. I'll do standard thinking. That's fine. All right. So literally nothing. I said nothing. I didn't have to click anything aside from once the skill is active. That's it.
Starting point is 00:30:51 Again, think of the benefit of using. standard operating procedures, templates, the concept of copy and paste, right? So now all these things that I might normally do in my day-to-day workflow, I can turn all of them into individual skills. I think the two biggest and best pieces of advice here as we wrap up our live demo portion, and then I'm going to tell you, I think, what are five great skill examples. Two big things. Number one, go find those long conversations that you had with chat, GPT, where you had to do
Starting point is 00:31:24 a lot of refining, a lot of back and forth, but you ultimately got the end deliverable, exactly how you wanted it or needed it. Go back and turn that into a skill. Take your time in doing it, right? And then iterate, use it once or twice. Look at the chain of thought, update it. Secondly, do exactly what I just showed you, right?
Starting point is 00:31:42 Say, hey, based on my chat history, based on my memories, or just upload files about yourself, based on all this, recommend 10 different skills that could end up saving me time or help me produce something more valuable to my clients, my potential customers, et cetera. So we'll see if this one worked now that I talked it up.
Starting point is 00:32:01 There we go. All right. So I pasted this long transcript, and it said right here, it said it used the Jordan hot take extractor skill. All right, and I can go in and view that, edit it,
Starting point is 00:32:13 download it, right? Because now I can take this elsewhere if I wanted to, but it did a good job. So it got my core stance here. It got my ranked hot takes. You know, one of them was most AI layoffs are PR. My second hot take there was Wall Street rewards AI layoffs.
Starting point is 00:32:31 Absolutely, right? So a good way to kind of get my opinions or my hot takes straight out. And then it got three additional episode angles. So if I wanted to do follow-up episodes and then quotable lines, all right. Hey, that's not bad. AI tools are the easy part, getting people to use them for real work. That's where companies fail. I agree with that.
Starting point is 00:32:52 Because I said it. All right. So there you go. Now you know the easiest way to go in and start using skills immediately. So let me quickly share three or no, five that I think could be helpful. All right. So one, automated executive briefings. So pulling in updates from connected company apps.
Starting point is 00:33:14 Remember, you can pull in information from multiple different apps. And then, you know, auto triggering that to perfectly format your morning update without needing to manually select a GPT or go into a project and, you know, connect a bunch of documents. Number two, budget versus actuals explainer. So pulling live data from connected spreadsheets to explain any variances in budgets, right? So again, versus saying uploading two documents, going through, explaining, here's what I want you to do.
Starting point is 00:33:42 Turn it into a skill. Third, marketing performance summaries. So drafting a weekly performance digest with insights and recommended actions. and this allows you to combine multiple analytics apps into one flawless, repeatable reporting workflow. Next, here's a great one, contract review summaries. Let's be honest, not everyone has the same skill sets, right? That's why I think the example of templates are great. Because think of like a PowerPoint template.
Starting point is 00:34:09 Not everyone is a good designer, but if you are a good designer, you can give someone the ability to design something that looks fairly good. Right. So in this example, a contract review summary, if your company has a strict process when it comes to legal review and you have all the documents, you have the examples, you have the SOPs, well, turn that into a skill because then you at least give people the ability to quickly understand maybe a contract or a service agreement or to be able to know where you can start to redline something without having to have all those skills. And then last but not least, automated operations reporting. So that. you know, pulling in operational KPIs from sheets, draft a weekly operations update, and that can kind of function as a strict SOP, ensuring that no section of the mandatory report is ever skipped or forgotten. So again, think, what do your people do over and over and over? And maybe there are some GPTs that work great, that you should probably turn into skills.
Starting point is 00:35:10 Maybe there's GPTs that laughed a little bit, that failed because it didn't have a certain feature or functionality that you. now have within skills or projects, right? Maybe there's something in projects, but you're like, ah, it's a little too limited. It didn't work really how we needed it to. So now I think you have a new tool. And sometimes it's not always the big release, right? It's not always going from, you know, GPT to 5 to GPD 5.5.
Starting point is 00:35:41 That's going to move the needle. Sometimes it is these, quote, unquote, smaller features that maybe you didn't even know about. right that can really help your company or your department turn a corner and work smarter, work more efficiently, and hopefully even create more valuable outcomes that can really push your company forward. All right. Was this one helpful, y'all? Putting AI to work on Wednesday.
Starting point is 00:36:06 So yeah, maybe you tune in once a week. We did change things up a little bit, FYI, and I'll just explain that here quickly. So Mondays, we bring you the AI news that matters. on Wednesdays, we usually go deep with one new feature or if there is a big model release, we'll do that. And then on Fridays, we're kind of starting to this new thing called feature Fridays because what I realized, it's very tough to choose that one thing to focus on on Wednesdays. And a lot of the most useful updates really just become a literal one second bullet point in our Monday podcast.
Starting point is 00:36:37 So that's kind of the lineup. And then Tuesday Thursdays, right, we'll rotate the shows, but doing a lot of the start here series and some guest interviews. here and there as well. So I hope this was helpful. If so, tell someone about it. Tell someone on your team. And you're like, yo, we need to start using these skills.
Starting point is 00:36:53 I just broke down the easiest and the simplest way for you to start using these today, giving you and your company the right information on how to start using skills as reusable templates today. So when you're done, listening this and sharing this with your friends, please, if you're listening on the podcast, please make sure to subscribe, then go to Your EverydayAI.com. Sign up for that free daily newsletter. Thanks for tuning in. We'll see back tomorrow and every day for more Everyday AI.
Starting point is 00:37:22 Thanks y'all. Meet Firefly AI assistant. Now live in Adobe Firefly, the Allman One Creative AI Studio. Just describe what you want to create in your own words and the assistant handles the rest, orchestrating multi-step workflows across Adobe Creative Cloud apps, including Photoshop, Premiere Express, and more in one conversational interface. You direct the outcome while the assistant accelerates. execution. Stand control with the ability to step in and refine at any time. See it today at
Starting point is 00:37:57 firefly.adobie.com. And that's a wrap for today's edition of Everyday AI. Thanks for joining us. If you enjoyed this episode, please subscribe and leave us a rating. It helps keep us going. For a little more AI magic, visit Your EverydayAI.com and sign up to our daily newsletter so you don't get left behind. Go break some barriers and we'll see you next time.

There aren't comments yet for this episode. Click on any sentence in the transcript to leave a comment.