🚀 Add to Chrome – It’s Free - YouTube Summarizer
Category: AI Tools
Tags: AIAutomationGoogleToolsWorkflow
Entities: CCDiscoGemini 3.0GoogleGoogle Workspace StudioMixboardNano BananaOpalPomelliSay What You SeeStitchZapier
00:00
So, Google might have been late to the air party, but they absolutely crushed 2025 with Gemini 3.0 and Gemini 3.0 Flash. Not only they have one of the best air models, it's the fastest, it's the cheapest.
In 2025, they dropped Nano Banana, Nano Banana Pro, which is the
00:16
best air image model on the planet, and also their Air Video model V3.1, which absolutely crushes. But on top of that, they launched another 25 air tools that you've never heard about.
And all these tools are built as experiments as part of Google map labs project. Actually you've probably heard of one notebook
00:32
alarm. But out of those 2025 tools they have dropped these eight tools are my absolute favorite and are super useful.
So let's dive into them. So first we have Google Workspace Studio.
So basically this is a tool by Google to build a workflows without any coding. So
00:48
if you just go to studio.workspace.google.com or google.com you should be able to see a dashboard like this where you can either just describe what workflow you want to create or you can just scroll down and you will see bunch of pre-built templates for example I can just build a workflow like every time I get a customer email with a negative sentiment
01:03
please label that email as emotional damage and use my customer email gem to create a draft response for that email so now when I just say create workspace studio will automatically build that AI workflow for me one of the biggest differences between this AI automation builder compared to everybody else is
01:20
ability to use your custom gems within the workflow and it's only possible because the whole platform is so native to Google Workspace. Now here we get to see the visual workflow builder.
There's a starter step which is when I get an email when I click on it I can define the conditions more if I want to fine-tune or really refine the
01:36
nitty-gritty of the steps. For example, should it run on a specific emails like from a certain email address or certain words it should have or is it going to be running on all the emails?
Then step two is going to be like define which is our action. Here I can type the prompt.
Go through my emails and see if the sentiment is negative. So then we have
01:52
like a workflow that says like check if step two if the decision is true. Now we have some of these follow-up steps.
So step one is going to be adding labels. Step two is going to be asking Gemini to now draft the response.
So a correction. So when I first tested this tool, we were able to actually build like use a
02:08
gem that was created within the workspace. But today I'm unable to find it.
I've checked my other previously built workflows as well and I'm unable to find the gem within those workflows as well. I wonder for some reason they actually deprecated that feature for now which is a bit of a bummer but don't worry I'm going to show you exactly if
02:24
you do want to use your gem how you can do so. Then we have a step which is going to be draft response or you can add any other substep you want.
So here we have an options. We can ask Gemini.
We can use some AI skills which is recapping emails, extracting, summarizing, some tools like filter or
02:39
we have some access to some tools within Gmail, chat, sheets, drive, etc. But this is very limited.
We only have access to apps within Google Suite and like we just noticed we weren't really able to call the gem. But again, if you want to build an AI workflow just like this, but you want to talk to other
02:55
apps, you want it to be have more control. Maybe you don't want to use Gemini, you want to use chat GBT or Collo or Plexity or any other AI tools or different business applications.
My go-to is Zapier. For example, this is my workflow for newsletter.
It's based on a schedule. So, every day at 3:30 p.m.,
03:11
it's going to start a conversation with Gemini, which has my system prompt to go research the news and updates, research the new tools, updates, and workflows. Then it's going to send this conversation again to Gemini, which is going to turn that into a newsletter draft based on our writing style.
Then it's going to use a Gemini function
03:27
again, but this time it's going to be generating an image. So we're going to be using Nano Banana as our model, which is going to be turning our news into like an abstract style newsletter cover.
Then it's going to go into my Air Table and it's going to automatically update all those fields. So my team can take
03:42
all this content now and turn that into a research. I've got a whole bunch of tutorials.
I have a whole playlist on Zapier AI workflows and automations. So feel free to check those out if you're interested in full tutorial on this.
But we're going to move on to other secret tools built by Google. And the next tool is OPPO.
So this is a tool that helps
03:58
you build like mini apps. The way I think about these mini apps is like a version of like custom GPT or Gemini gem combined with a little bit of automation and cloud artifacts because everything we build has a lot more visual layer to play with it.
For example, this is like a Opal mini app for book recommendation.
04:15
Here we have an example of you start with describing what you want and then it's going to go through a bunch of steps which are basically different prompts that are all strung together and then our output is going to be essentially like a web page. So which is where we could like share this with anyone.
It's like a mini app. So let's say if we go to preview and we start
04:31
describe the kind of book there's a typo here kind of book you're looking for something non-fiction but about impact of AI on civilization if I can spell it right. Now we have this mini app that's working.
I mean behind the scenes we can see it's just going through all these different steps of prompts gathering data talking to like different tools
04:48
within Google and it's generating the output but here we have a few book recommendations life 3.0 being human in age of artificial intelligence AI 2041 10 visions of our future super intelligence the age of AI I think I actually have a copy of this one actually human compatible so this is
05:03
Google Opal okay let's move on this one has improved so much over the past 6 months I think especially after the Gemini 3.0 Odrop and they integrated that into this tool. So this one is called stitch.
It's essentially a way like to replace a junior UX UI designer. And I'm not exaggerating when I say that
05:20
because I'm going to show you the quality of results you get. So it's not really like an app builder per se, but it will build you the prototype and it's going to build the visual layout for any idea that you may have.
For example, we can just say make me an app for discovering mango activities and getaways. Focus on showcasing destinations, musical events, and easy trip planning.
Now here have an option
05:36
to like build this for web or for mobile. So let's just say if we do for web.
So we have a model which is 3.0 probably the best model that Gemini has right now. So let's just say what it builds us.
So here on the left we can say to design a comprehensive app for discovering Vancouver activities and getaways. We can start with the following screens.
So we have Vancouver
05:53
discover dashboard. We're going to get Vancouver destination list musical events calendar planner interface.
Shall we proceed with all these screens? And I'm just going to say like yes.
Let's do them all. Or I can just describe what I want in this chat interface.
Looks like we have our first app here. If we zoom
06:10
in here, it looks pretty modern. We don't see that purple that every single AI coded websites usually gets.
This is the Capalano suspension braid. It's Gaston Stream Clock.
It's pretty accurate, I would say. You know, being a Vancouver right here.
We got the perfect skyline here for the destination list.
06:25
We have top destinations. I would say it's a pretty good design.
So, if you are trying to like work with a junior designer, I think this is the design that freelancer would create for you. Here we have a card style list for different events.
Here we have like some of the screen for like the trip planning interface. Honestly, I'm a fan of this design.
And then there's a cool couple
06:41
features I want to show you. So, first is let's say we just go here.
If we go to generate here, we have an option for just variations. So, maybe you're not happy with this.
Again, it's a visual layout. You can just do variations and it's going to generate multiple variations of this landing page.
So, you can pick whichever you want. But what I like is this predictive heat map which
06:57
is great for sales pages. So essentially what it's going to do is it's going to create like you know simulation of where the user attention is going to go on the page.
Especially if you're trying to do e-commerce, you know, anything that's booking related, anything that's commerce related, it's great to do the visual like the predictive heat map. So
07:14
you can see if like the attention is going to be on the button to sign up. So exactly what action you want the user to take.
If like the exact if the red if the red heat area is not there, you want to change that design up. So again, it's just a prediction.
So just AI is kind of
07:30
using probably some design psychology to see where the user attention is going to be. Usually it's like more like an F shape anyway.
But it's a kind of quick uh sanity check to see if your design is going to be good enough or not if you are planning on using this tool. And then I think the other option I let's see if we can get this with here.
If I
07:45
go to more, I have an option to view the code. I can export.
So I can export this as like a zip file. I can export this to Jules, which is like their tool for V coding, which is great.
you can just quickly turn your designs by code at all. There was an option somewhere to like do that as a prototype as well.
So
08:01
let's just say we select these three screens. Now if we go here prototype now I'm going to get more like a clickable version of prototype which is again great for just like getting some customer feedback or just testing internally with the team how the flow looks like.
Okay, we have our protype. Let's see.
So if I go here and interact
08:17
with a protype. Now I'm interacting with the protype.
Let's just say I click sign in. I can type here next getaway Vancouver search.
Some of the buttons are not working as well, but we have the events page working. Maybe actually we didn't really have the screens yet.
Maybe that's why. So, we have the events
08:33
page working. I mean, I like the hover it's getting.
So, we're actually able to visualize what the hover looks like. So, the three screens that we had in our prototype was going to be the main page, the musical events calendar, and then we had the trip planning interface.
So, if we go here, we are right now on our
08:48
events page. And let's see if we can navigate to our trip planning interface.
So trip planner. If we go here, we're actually able to get this interface now.
So we're able to like create interactivity among the three screens that we did select. And I like like the
09:04
little micro animations that we have in the design. So here it's like highlights when I drag to upload any activity here.
Drag any activities. There's a text field that I can play with to visualize.
Delete is working. That's cool.
So that's like a quick prototype we were able to build that's interactive using the platform search with google.com. All
09:20
right, let's move on. So next one, you may have heard about this one briefly on like you know through show form video.
It's called Pomelli. So this is a tool they build to rival like Canva essentially which is essentially for like small teams or people who don't have big like or any marketing teams to
09:36
build on brand marketing assets. Essentially the way it works is that you upload your website URL whatever that is.
Then it's going to extract your business DNA, which means the colors you use, the fonts you use, and all the images that you already have on the website. Now you just have to type your
09:51
prompt, what sort of marketing campaign you want, and it's going to generate some assets for you like your ads, social media assets like images, photos, different aspect ratios. That's going to be on brand using the images from your website.
So if you don't have a marketing team, you don't do any marketing at all, Pomelli is is a great
10:07
tool. And they recently announced animation as well.
So, not just the static designs using VO3.1. It can also create some mini animations of your static of your assets you've been creating.
It actually requires less skill than learning how to use Canva, but can create some quality assets for
10:23
you. But sometimes there is some consumer feedback, it just doesn't work as well.
So, maybe if your brand has a lot more assets on the website, it's going to work well. But if your website is a bit limited, it might fail for you.
The next two ones are going to be the tools that are currently on weight list. They were just announced last week.
10:39
First is called disco. So disco is going to be an interesting one.
It's planning on changing the way we use our browsers. For example, if you're planning a trip to a saka, uh you have like multiple tabs open, you know, things to do, tickets, pricing, and all that stuff,
10:54
recommendations. Now, when you use Disco, it's going to turn all those different tabs into one interactive app automatically.
Maybe you're trying to learn a new topic. Maybe you're trying to learn like anthropy and you have bunch of different papers open, different tabs open, YouTube videos open, like all the content can be turned
11:10
into one interactive experience in a single page that they're going to call like a gen tab. So your experience can be a lot more interactive by accumulating information across multiple tabs.
So which is why I say it's like a newer version of how we use the internet. Sadly, I don't have a personal
11:25
experience using it just yet. I don't have access to it.
It's on weight list. But once I do, I will do a review video of it.
Let me know if you do want me to do so. The next one which is on weight list is called CC which is going to make working with your emails a lot more easier.
Even with Gemini, I feel like searching exactly a certain specific
11:42
thing in your email is a lot harder. So hopefully they'll fix that soon as well.
But essentially what this does is it gives everyone a personal secretary. So job offer secretary used to be like telling you exactly what you need to know when you walk into the office every morning.
But with CC it works the same way. It's going to go through different
11:58
emails and work files and it's going to tell you give you a summary of exactly what you need to do today. What should be top of mind?
What's on your calendar? What things you should be like just be aware of that might be relevant to the work you need to do today.
On top of that, just like any secretary, you'll be like, "Hey, put this on my calendar."
12:13
Like every time you have like an action item or something, you can just like CC your assistant and it will just add those things to your task or calendar automatically for you. So, the idea of CC is to give everyone their own personal secretary, which would be kind of cool if it works as well as it is advertised.
Again, it's a second tool,
12:29
which I said I don't have experience playing with just yet. It's also wait list, but I think they're going to be rolling it out in the new year because they just announced this like last week.
All right, moving on. So, this one actually helps you learn how to write prompts for images, but in a gamification way.
So, it's called say what you see. So, essentially here my
12:46
job is to look at the picture and just say what I see. So the idea is that by accurately describing the image I will learn how to like prompt.
It's kind of like you know reverse engineering. So I'll be better able to communicate with AI what I want to create.
This one's going to be hard one but let's just say
13:02
a flower pot with roses but it is embroidered. So, what it's going to do is it's going to take my prompt and it's going to generate the image based on my prompt and then compare the similarity between what it was and what my image was and it's going to show me the prompt that I was supposed to write.
An
13:18
embroidery of a vase of red roses and I said a flower pot. Oh, yeah.
I mean vase and pot that would make sense. So, the same way I can go next and I can just like keep doing this over and over again here.
Now, I don't see what the original prompt is and I'm supposed to recreate this image to 100% match. As you can
13:34
see, like the more imaginative you are, the more creative you are on explaining these things, the better you're going to be. If you want to break into a image space, I think it's a great great tool to like learn and practice.
Just go ham. Like make it a Dualingo thing.
Well, that's a great idea. You can like build this like a Dualingo version as an app
13:50
and maybe try to monetize it. That could be cool.
Next one, I think this is our last one. It's called Mixboard.
If you generate images, this is like more like a visual brainstorming tool. So here I have like a whole canvas where I have bunch of different images.
is I can do a whole bunch of different things. I have a few assets that are there.
And now I
14:06
can just turn those assets and either I can remix it. For example, this is a style.
There's a plate. Hey, replace the pattern on the plate.
Boom. Done.
Or I could just be like I can pick the pattern, turn this into a mockup of a t-shirt and also a keychain. Very random, but let's see what happens.
But
14:22
essentially, like the idea is that mixport is to like mix and match different assets, create multiple different visuals, brainstorm. Maybe you're experimenting with certain like material styles or certain design aspect.
You have a design aspect, you can turn that into like multiple product images. You can do multiple products.
Here for example, we have now this as a
14:39
keychain and also as a t-shirt. So you can see like how whatever I'm trying to brainstorm, I can start experimenting within the mixport space like really helps with brainstorming different visual ideas especially for branding and marketing work.
So yeah, these were the top tools I think that are quite hidden, quite secret. Google Apps project unlike
14:56
Notebook LM which just took off, but I think these are going to be super relevant in 2026. And I'm glad you're now aware of these tools.
Also, it's good to know because some people might just like package these tools and apps and trying to sell it to you. But now you know there's a better or freer version of these tools that you could just be using instead.
Let me know what do you want to see in the new video. If
15:12
this was useful, give a thumbs up, subscribe, comment, and I'll see you in the next video. your GPA officer signing