July 26, 2026 • 13 mins
The AI world has has done a spectacularly bad job of naming things. Agents, GPTs, gems, notebooks, projects, skills - you'll find a completely different word for what is essentially the same kind of tool. Agents in one place, GPTs in another, gems somewhere else. Different names, similar ideas - and no handbook to tell you that.
But the tools themselves are not that complicated. Once you know what each one is built to do, knowing which to reach for becomes straightforward.
In this How I AI episode, Neo and I cut through the jargon and map out what Agents, Projects, and Skills actually are, how they differ from each other, and which one to reach for depending on what you need to do.
What you'll learn:
- Why agents, GPTs, and gems are the same kind of thing (just with different names)
- When a project is the right choice over an agent
- What makes a skill different from a long prompt
- Practical AI tools for productivity and focus
- Real-world AI workflows used by high performers
- How to use AI at work without burning out
- Smart shortcuts for managing time and mental load
Connect with Neo Aplin on LinkedIn (https://www.linkedin.com/in/neoaplin/) and via inventium.ai (https://inventium.ai), where he leads Inventium's AI training and upskilling work with organisations and teams.
How I AI with Dr Amantha Imber and Neo Aplin explores how high performers are using AI at work to boost productivity, make better decisions and reduce overwhelm.
Want more practical ways to put AI to work? You can now follow How I AI separately on Spotify or Apple Podcasts so you never miss an episode.
My latest book The Energy Game is out now. You can order a copy here: https://amzn.to/48ID29M
Connect with me on the socials: Linkedin (https://www.linkedin.com/in/amanthaimber)
Instagram (https://www.instagram.com/amanthai)
If you are looking for more tips to improve the way you work and live, I write a weekly newsletter where I share practical and simple to apply tips to improve your life. You can sign up for that at https://amantha.substack.com/
V
Listen
Watch
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:00):
If you've ever sat there wondering what the difference is
between an agent, a project, and a skill, and whether
you're supposed to be using a GPT, JEM, a notebook
or something else entirely, you are not alone. The AI
industry has done a spectacularly bad job of naming any
of this, and it makes getting started feel harder than
(00:21):
it should. In this episode, Neo and I try to
sort it out properly. You will learn what an agent
actually is and when to reach for one, why projects
shine when you've got a pile of documents you want
to ask questions about, and how skills work like a
recipe your whole system can follow. By the end of
(00:42):
this episode, you will know exactly which of the three
to use and when, with no more guessing involved. Welcome
to how IAI with me, Doctor Amantha Imba and Neo Applin,
head of Inventium AI. Each episode we share one practical
(01:05):
way to use AI better at work and in life.
No fluff, no tech jargon, just things you can use
straight away, Neo. As we often complain about on how
I AI, there are so many different words and terms
for things, and for someone that is not deep in AI,
every day like us, it can be really confusing, and
(01:28):
so today I want to unpack water skills versus water
projects versus what are agents slash GPTs slash gems. So
let's start with agents, which in Copilot we're talking about agents,
in Chat gpt we're talking about GPTs, and in Gemini
we are talking about gems. What are these things neo
(01:51):
and when should we use them?
Speaker 2 (01:53):
Apologies on behalf of the whole AI industry. They're horrendous
at the naming on this one. So we are talking
about this. It's ish kind of thing. And really the
Copilot agent Chachapite, GPTs and Gemini gems I might just
call them in air quotes agents. These ones are ais
with a job. So you get them to do something.
(02:14):
You give them a task, which is generally some instructions.
You might give them a couple of templates or look
up files or something like that to be able to
answer your questions or to help do the task, and
they do that one job. There's a big asterisk here
in that it doesn't take forever to do the job.
It will not work for twenty four hours, it will
not wake up at three in the morning. You give
it something to do, it will do that thing and
(02:36):
we'll finish it within a minute or two generally, and
when it produces for you is generally up to about
like five pages worth of content. So if you've got
a spreadsheet with a thousand rows, it is not the
task for you. But if you say, hey, I've got
a meeting transcript that I want you to clean up
using this template, it is absolutely the job for you
to use. So that is an agent, copilot, agent, chat TOPTGBT, Gemini, GM,
(03:01):
an AI with a job.
Speaker 1 (03:02):
Okay, so, Neo, what are maybe two or three of
the most common or the agents that you used most
often in your work life?
Speaker 2 (03:15):
Wow? Okay, I am not usual because I'm always creating
agents and things like that. Where I think most people
will probably use them would be within their job. So
what it comes down to is you've got to figure
out what you're doing to figure out whether you need
an agent to help you with that. So I have
a researcher agent that helps me to specify what I'm
(03:36):
researching and ask me questions about that, so I can
then give that to a deep research tool. So I
get a really good research agent. Why because I need
to keep up to date with things, and I'm always
learning things, and so researching is very much part of
my job. A lot of people would probably have meeting
minute cleanup agents or gipitis and gems, and what that'll
do is you'll take maybe your meeting transcript, maybe take
(03:57):
your agenda, and it'll then format it into a report
or an output which might be his the items and
maybe in a table for who's doing what when and
things like that, and maybe you can even call out
these were the things that we're on the agenda that
weren't talked about. So those kind of things, meeting agents
and things like that. I've built a bunch for HR people.
So it's things like maybe reviewing job ads before you
(04:22):
send them out, or it might be a competency framework
and things like that which are very much detailed stuff
that you do that he needs a lot of checking
and things like that. Agent's GPT's gems are awesome for that.
Speaker 1 (04:34):
Let's now talk about projects. So I know in Claude
and chat JPT they are called projects. What is the
equivalent in Gemini and co pilots.
Speaker 2 (04:47):
Not an equivalent in Gemini, They've just got gems in Gemini.
In co pilot. They've kind of sort of got of equivalent. Maybe. Okay,
it's called notebooks. It's actually just gone through a bit
of a reach. Actually looks a little bit closer to
a project in the others, So I'd say, yeah, notebooks
are probably equivalent. But projects and notebooks are similar kind
(05:08):
of things, and they're also kind of similar to an
agent GPT and GEM in that it takes your instructions
and also particularly take a board. We'll take a lot
of your files so you can answer questions about those files.
That's how they're engineered.
Speaker 1 (05:21):
So when should we use a project as distinct from
an agent?
Speaker 2 (05:26):
Okay, it depends on the tool you're using. So if
you're in let's say chat GPT, because this makes it
nice and complicated because you've got GPTs, you've also got
projects and they're also got skills, which will cover in
a second. So I would use a GPT when I've
got a single task, a single job that I need
to do again and again and again. I would use
(05:47):
a project if I've got a lot of knowledge, a
lot of information there that I need to ask questions about,
or maybe i've got a topic like a real project
that I'm actually need to do lots of work about
because one of the great things about projects is yes,
it'll lots of documents, but also groups your chats about
those topics together. So you might have a maybe you're
(06:08):
implementing a new CRM, so you might have the new
CRM projects, and you'll have all the files about that
new CRM and all the chats you have about that
new CRM. Well then get grouped together, which is kind
of cool in one project. So I use projects for
things that you work on. So I've got a project
for Inventium, which has a lot of products and services
and things like that. You might have one for like
a standard operating procedures. You might have one as like
(06:29):
an onboarding one. You might have one standard Q and
a one up to you what you need.
Speaker 1 (06:34):
So am I right in saying that with a project,
projects shine where there's heaps of context that you want
to have the AI know about you to do a
really good job in what you're asking it to do.
Speaker 2 (06:47):
Absolutely. Yeah. So the more context, more backgrounds, it's very
much a document heavy, conversation heavy thing rather than a
job heavy thing. So if I'm asked getting questions as
answered from documents from knowledge. Absolutely our create a project.
Speaker 1 (07:02):
Okay, finally, let's talk about skills now, just to get
the language right. In Claude, we have skills in chat GPT,
we have skills in co pilot. We also have in.
Speaker 2 (07:16):
Copilot Cowork, we have skills. Believe it or not. So
this is it's come out in the last few weeks,
so you may see it on your copilot screen. So
a couple of caveats. This is only for company users.
So this is the Microsoft three six y five copilot
users and your company may or turn it on, they
(07:37):
may turn it off. And what it is, it's part
of the Cowork type thing. It very much feels like
claud Cowork. They've used a lot of that technology there.
And skills in copilot can only be used in copilot Cowork.
They cannot be used in regular copilot.
Speaker 1 (07:53):
Great, just to make everything more confusing. And do we
have an equivalent in Gemini.
Speaker 2 (07:58):
No, No skills in Gemini, although I wouldn't be surprised
if some stage it comes out. It's become an open standard.
And this is where I love the word skills, because
we're talking about the same kind of thing. In we
at least three of the four main tools. And so
skills are actually really good and they're like a reusable method,
so it's something that you can do again and again
and again. So what you will find, maybe I'll give
(08:22):
you an example of how these things kind of fit together.
So I want you to imagine like a restaurant kitchen.
So I love going to a restaurants where you can
actually see the kitchen and maybe you look at the
bar and watch people do their jobs, and I find
it fascinating because they're awesome. And what you find in
a restaurant kitchen is you'll have specialists in different areas.
You might have someone at the grill, they're the girl master,
and maybe you'll have the pastry chef and they're sitting
(08:44):
over there doing the pastry stuff. And generally what you
find is each of these cooks, these chefs just doing
one thing, if you'd like, that's pretty much like an
agent GPT in a gym. So they do one thing,
they know what they're doing, they do that thing really
really well. A skill is more like a recipe, and
the recipe is then how to do those things and
(09:06):
the steps you do to be able to do, you know,
making some pastry or to be able to grill a burger,
which is very different than maybe grilling some lamb chinks
or something like that. And so the skill is more
like a recipe that can be used by you or
pretty much anyone in the system.
Speaker 1 (09:24):
I think one of the ways I think about skills
is like standard operating procedures. So where it's like do
this than this than this, and what I know I
think you taught me this is that where a skill
is better than a long prompt is that adherence is better. So,
particularly in some of the AI, is if I paste
(09:46):
in a really long prompt that say, you know, could
also be a skill. Sometimes the AI will be a
little bit hard of hearing and it might miss some
of the steps. But in a skill, am I correct
in saying that the adherence to each and every step
in that recipe or standard operating procedure will be adhered to.
Speaker 2 (10:09):
With anything, there's a massive exception because you can write
good skills and bad skills and things like that. Skills
are less often less directive. They're less about do this,
this and this and this as far as the minutia
of it all, but they do have overall steps. So
you need to include these things and things like that.
Skills work really well with goals, So here's what the
(10:29):
outcome is supposed to be, here's some templates of what
the outcome needs to be, and things like that. Skills
often can do other things as well. So they might
be able to invoke a Python script. In other words,
it can do things with data and things like that.
It could do things with pulling together files in different
ways and things like that. So skills are very very capable,
and they can then refer to other skills as well,
(10:52):
so one skill can effectively hand off to another one
or recommend that the next step is the next skill,
and things like that. So yeah, skills can be very
valuable and I prefer using skills over massive prompts absolutely.
Speaker 1 (11:05):
So if we were to summarize the difference between agents, projects,
and skills, how would you do that? In a nutshell,
it's more.
Speaker 2 (11:16):
About what tools you've got and then how you want
to solve a work problem. So we'll start with projects first.
Projects are great at solving the problem on I've got
to get information from these documents, from these sources, and
it is excellent at doing those. And you can also
curate what goes into the documents and the sources and
things like that. So I use projects very much. For
(11:37):
like I've got an an inventium and personal briefing documents
about what I'm doing and what we're working on and
things like that, and they're awesome for here's the procedures
and the policies and things like that, so awesome for those.
I would use a copilt agent, CHPTGPT or a Gemini
gem to be able to do a single task. So
I've got a meeting, I need to clean it up.
(11:59):
I've got, As I said research before is really good
for those kinds of things. I even use it for
helping me to refine our language and prompts and things
like that. So particularly with language, I've got like a
Ventium style skill which makes things sound like our brand style.
So so not skilled agent, I should say to do that.
Most of these agents, you could also build them as
a skill as well. And so if you've got a
(12:21):
tool like CHPT that has GPTs and skills, the more
modern way of doing it is absolutely built as a skill.
So a skill is great. If you've got like a
step based thing, you've got some guides, you've got some
templates and things like that. The great thing about the
skill is it can be used pretty much throughout the area.
So it's almost like you've got a head chef who
can run around the kitchen and use any of the recipes,
(12:42):
any of the skills that they want to, as opposed
to like an agent, GPT a gem, which is just
a single cook that can only cook one thing, one recipe. Okay,
I love that.
Speaker 1 (12:52):
I hope that this has been useful for people who
are just feeling a little bit more overwhelmed by all
the different to and ways to get the most out
of AI. Hopefully now you are clearer on what is
a skill, what is a project, and what is an
agent and when to use each of those three very
distinct methods in AI. How I AI was hosted by
(13:15):
me Amantha Imber and Neo Applan. A big thank you
to Martin Imber who does our sound editing, and Jim
Rubio for production support, and thank you to John Kilby
who composed the theme music.
If you've ever sat there wondering what the difference is
between an agent, a project, and a skill, and whether
you're supposed to be using a GPT, JEM, a notebook
or something else entirely, you are not alone. The AI
industry has done a spectacularly bad job of naming any
of this, and it makes getting started feel harder than
(00:21):
it should. In this episode, Neo and I try to
sort it out properly. You will learn what an agent
actually is and when to reach for one, why projects
shine when you've got a pile of documents you want
to ask questions about, and how skills work like a
recipe your whole system can follow. By the end of
(00:42):
this episode, you will know exactly which of the three
to use and when, with no more guessing involved. Welcome
to how IAI with me, Doctor Amantha Imba and Neo Applin,
head of Inventium AI. Each episode we share one practical
(01:05):
way to use AI better at work and in life.
No fluff, no tech jargon, just things you can use
straight away, Neo. As we often complain about on how
I AI, there are so many different words and terms
for things, and for someone that is not deep in AI,
every day like us, it can be really confusing, and
(01:28):
so today I want to unpack water skills versus water
projects versus what are agents slash GPTs slash gems. So
let's start with agents, which in Copilot we're talking about agents,
in Chat gpt we're talking about GPTs, and in Gemini
we are talking about gems. What are these things neo
(01:51):
and when should we use them?
Speaker 2 (01:53):
Apologies on behalf of the whole AI industry. They're horrendous
at the naming on this one. So we are talking
about this. It's ish kind of thing. And really the
Copilot agent Chachapite, GPTs and Gemini gems I might just
call them in air quotes agents. These ones are ais
with a job. So you get them to do something.
(02:14):
You give them a task, which is generally some instructions.
You might give them a couple of templates or look
up files or something like that to be able to
answer your questions or to help do the task, and
they do that one job. There's a big asterisk here
in that it doesn't take forever to do the job.
It will not work for twenty four hours, it will
not wake up at three in the morning. You give
it something to do, it will do that thing and
(02:36):
we'll finish it within a minute or two generally, and
when it produces for you is generally up to about
like five pages worth of content. So if you've got
a spreadsheet with a thousand rows, it is not the
task for you. But if you say, hey, I've got
a meeting transcript that I want you to clean up
using this template, it is absolutely the job for you
to use. So that is an agent, copilot, agent, chat TOPTGBT, Gemini, GM,
(03:01):
an AI with a job.
Speaker 1 (03:02):
Okay, so, Neo, what are maybe two or three of
the most common or the agents that you used most
often in your work life?
Speaker 2 (03:15):
Wow? Okay, I am not usual because I'm always creating
agents and things like that. Where I think most people
will probably use them would be within their job. So
what it comes down to is you've got to figure
out what you're doing to figure out whether you need
an agent to help you with that. So I have
a researcher agent that helps me to specify what I'm
(03:36):
researching and ask me questions about that, so I can
then give that to a deep research tool. So I
get a really good research agent. Why because I need
to keep up to date with things, and I'm always
learning things, and so researching is very much part of
my job. A lot of people would probably have meeting
minute cleanup agents or gipitis and gems, and what that'll
do is you'll take maybe your meeting transcript, maybe take
(03:57):
your agenda, and it'll then format it into a report
or an output which might be his the items and
maybe in a table for who's doing what when and
things like that, and maybe you can even call out
these were the things that we're on the agenda that
weren't talked about. So those kind of things, meeting agents
and things like that. I've built a bunch for HR people.
So it's things like maybe reviewing job ads before you
(04:22):
send them out, or it might be a competency framework
and things like that which are very much detailed stuff
that you do that he needs a lot of checking
and things like that. Agent's GPT's gems are awesome for that.
Speaker 1 (04:34):
Let's now talk about projects. So I know in Claude
and chat JPT they are called projects. What is the
equivalent in Gemini and co pilots.
Speaker 2 (04:47):
Not an equivalent in Gemini, They've just got gems in Gemini.
In co pilot. They've kind of sort of got of equivalent. Maybe. Okay,
it's called notebooks. It's actually just gone through a bit
of a reach. Actually looks a little bit closer to
a project in the others, So I'd say, yeah, notebooks
are probably equivalent. But projects and notebooks are similar kind
(05:08):
of things, and they're also kind of similar to an
agent GPT and GEM in that it takes your instructions
and also particularly take a board. We'll take a lot
of your files so you can answer questions about those files.
That's how they're engineered.
Speaker 1 (05:21):
So when should we use a project as distinct from
an agent?
Speaker 2 (05:26):
Okay, it depends on the tool you're using. So if
you're in let's say chat GPT, because this makes it
nice and complicated because you've got GPTs, you've also got
projects and they're also got skills, which will cover in
a second. So I would use a GPT when I've
got a single task, a single job that I need
to do again and again and again. I would use
(05:47):
a project if I've got a lot of knowledge, a
lot of information there that I need to ask questions about,
or maybe i've got a topic like a real project
that I'm actually need to do lots of work about
because one of the great things about projects is yes,
it'll lots of documents, but also groups your chats about
those topics together. So you might have a maybe you're
(06:08):
implementing a new CRM, so you might have the new
CRM projects, and you'll have all the files about that
new CRM and all the chats you have about that
new CRM. Well then get grouped together, which is kind
of cool in one project. So I use projects for
things that you work on. So I've got a project
for Inventium, which has a lot of products and services
and things like that. You might have one for like
a standard operating procedures. You might have one as like
(06:29):
an onboarding one. You might have one standard Q and
a one up to you what you need.
Speaker 1 (06:34):
So am I right in saying that with a project,
projects shine where there's heaps of context that you want
to have the AI know about you to do a
really good job in what you're asking it to do.
Speaker 2 (06:47):
Absolutely. Yeah. So the more context, more backgrounds, it's very
much a document heavy, conversation heavy thing rather than a
job heavy thing. So if I'm asked getting questions as
answered from documents from knowledge. Absolutely our create a project.
Speaker 1 (07:02):
Okay, finally, let's talk about skills now, just to get
the language right. In Claude, we have skills in chat GPT,
we have skills in co pilot. We also have in.
Speaker 2 (07:16):
Copilot Cowork, we have skills. Believe it or not. So
this is it's come out in the last few weeks,
so you may see it on your copilot screen. So
a couple of caveats. This is only for company users.
So this is the Microsoft three six y five copilot
users and your company may or turn it on, they
(07:37):
may turn it off. And what it is, it's part
of the Cowork type thing. It very much feels like
claud Cowork. They've used a lot of that technology there.
And skills in copilot can only be used in copilot Cowork.
They cannot be used in regular copilot.
Speaker 1 (07:53):
Great, just to make everything more confusing. And do we
have an equivalent in Gemini.
Speaker 2 (07:58):
No, No skills in Gemini, although I wouldn't be surprised
if some stage it comes out. It's become an open standard.
And this is where I love the word skills, because
we're talking about the same kind of thing. In we
at least three of the four main tools. And so
skills are actually really good and they're like a reusable method,
so it's something that you can do again and again
and again. So what you will find, maybe I'll give
(08:22):
you an example of how these things kind of fit together.
So I want you to imagine like a restaurant kitchen.
So I love going to a restaurants where you can
actually see the kitchen and maybe you look at the
bar and watch people do their jobs, and I find
it fascinating because they're awesome. And what you find in
a restaurant kitchen is you'll have specialists in different areas.
You might have someone at the grill, they're the girl master,
and maybe you'll have the pastry chef and they're sitting
(08:44):
over there doing the pastry stuff. And generally what you
find is each of these cooks, these chefs just doing
one thing, if you'd like, that's pretty much like an
agent GPT in a gym. So they do one thing,
they know what they're doing, they do that thing really
really well. A skill is more like a recipe, and
the recipe is then how to do those things and
(09:06):
the steps you do to be able to do, you know,
making some pastry or to be able to grill a burger,
which is very different than maybe grilling some lamb chinks
or something like that. And so the skill is more
like a recipe that can be used by you or
pretty much anyone in the system.
Speaker 1 (09:24):
I think one of the ways I think about skills
is like standard operating procedures. So where it's like do
this than this than this, and what I know I
think you taught me this is that where a skill
is better than a long prompt is that adherence is better. So,
particularly in some of the AI, is if I paste
(09:46):
in a really long prompt that say, you know, could
also be a skill. Sometimes the AI will be a
little bit hard of hearing and it might miss some
of the steps. But in a skill, am I correct
in saying that the adherence to each and every step
in that recipe or standard operating procedure will be adhered to.
Speaker 2 (10:09):
With anything, there's a massive exception because you can write
good skills and bad skills and things like that. Skills
are less often less directive. They're less about do this,
this and this and this as far as the minutia
of it all, but they do have overall steps. So
you need to include these things and things like that.
Skills work really well with goals, So here's what the
(10:29):
outcome is supposed to be, here's some templates of what
the outcome needs to be, and things like that. Skills
often can do other things as well. So they might
be able to invoke a Python script. In other words,
it can do things with data and things like that.
It could do things with pulling together files in different
ways and things like that. So skills are very very capable,
and they can then refer to other skills as well,
(10:52):
so one skill can effectively hand off to another one
or recommend that the next step is the next skill,
and things like that. So yeah, skills can be very
valuable and I prefer using skills over massive prompts absolutely.
Speaker 1 (11:05):
So if we were to summarize the difference between agents, projects,
and skills, how would you do that? In a nutshell,
it's more.
Speaker 2 (11:16):
About what tools you've got and then how you want
to solve a work problem. So we'll start with projects first.
Projects are great at solving the problem on I've got
to get information from these documents, from these sources, and
it is excellent at doing those. And you can also
curate what goes into the documents and the sources and
things like that. So I use projects very much. For
(11:37):
like I've got an an inventium and personal briefing documents
about what I'm doing and what we're working on and
things like that, and they're awesome for here's the procedures
and the policies and things like that, so awesome for those.
I would use a copilt agent, CHPTGPT or a Gemini
gem to be able to do a single task. So
I've got a meeting, I need to clean it up.
(11:59):
I've got, As I said research before is really good
for those kinds of things. I even use it for
helping me to refine our language and prompts and things
like that. So particularly with language, I've got like a
Ventium style skill which makes things sound like our brand style.
So so not skilled agent, I should say to do that.
Most of these agents, you could also build them as
a skill as well. And so if you've got a
(12:21):
tool like CHPT that has GPTs and skills, the more
modern way of doing it is absolutely built as a skill.
So a skill is great. If you've got like a
step based thing, you've got some guides, you've got some
templates and things like that. The great thing about the
skill is it can be used pretty much throughout the area.
So it's almost like you've got a head chef who
can run around the kitchen and use any of the recipes,
(12:42):
any of the skills that they want to, as opposed
to like an agent, GPT a gem, which is just
a single cook that can only cook one thing, one recipe. Okay,
I love that.
Speaker 1 (12:52):
I hope that this has been useful for people who
are just feeling a little bit more overwhelmed by all
the different to and ways to get the most out
of AI. Hopefully now you are clearer on what is
a skill, what is a project, and what is an
agent and when to use each of those three very
distinct methods in AI. How I AI was hosted by
(13:15):
me Amantha Imber and Neo Applan. A big thank you
to Martin Imber who does our sound editing, and Jim
Rubio for production support, and thank you to John Kilby
who composed the theme music.
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
The Bobby Bones Show
Listen to 'The Bobby Bones Show' by downloading the daily full replay.