Buzz: New FREE AI Agent!

Today, I'm going to be running you through Buzz, which is a new workspace for human and agent teams. So it's kind of like an AI powered Slack. Super easy to use, very intuitive. And if you're finding like other AI agents, for example, like Hermes or OpenCore or a bit technical, this is actually much more easier to use.

So what you can see here is that we have a kind of like a Slack version with AI. And you've got these different channels. So over here, for example, we have blog creation. And what we've done is basically created separate channels for each of our AI agents, depending on what they're going to work on.

So if we have a look over here, we've got this section for blog creation. And then we said, create a new blog post about this tagged in the AI agent we've installed, and then it's replied to us. So we can reply and message back in real time. Also, what I like about this is it's just much easier to track your workflows and track previous conversations by the project.

So for example, if you're using like even Claude, Claude is pretty hard to track everything that's going on, as you can see right here inside Claude desktop, pretty hard to see everything and go back on recent conversations. With the setup inside Buzz, we've got separate workflows so that we can see the history of everything that we've done for that particular workflow. So for blog creation, we actually said, Fizz, create a new blog about this AI customer agent and publish it. And so Fizz is the AI agent that we have plugged into this.

So it comes built into Slack. Now, what's also pretty cool about this is we didn't have to give it any training, but it already has information on all of our contexts. So if you have API keys, for example, I've already given in the past Claude code access to my WordPress and also the API access token for WordPress, then you can see here that it can publish directly to our WordPress website. So when we said create a new blog post about this for this keyword, actually publish that article, as you can see on our website directly for that particular keyword, which is pretty awesome.

So it was pretty easy and simple. And not only that, but it used our custom workflow that already trained Claude on. So the benefit of this is that number one, you can plug Claude into Buzz, which you can't do with Home is Agent. You can't use Claude with Home is Agent or any other agent for like, for example, OpenClaude and stuff like that.

And then number two, it's already pre-trained on all of your contexts. So all the time you spent training Claude or Codex in the past, it already has those workflows pre-built in. So you can pull them in as soon as you start using Buzz. And then finally as well, it's a lot easier to use, just like super easy UI, which makes it way easier.

And additionally, so far Buzz has been free. So I think it's an open source project. I saw it trained on GitHub. That's why I decided to show it, but yeah, pretty nice so far.

Now you also get like three different agents. And another benefit of this is that you can actually add your team to the workflows as well. So if we click on, for example, our settings in the bottom left here. So if we click on settings on our profile, we can then actually add people to our team.

So if we go to the invite section here, we can invite people to our community, which I think is really cool because let's say, for example, you have a bunch of friends that do AI agents. You can actually invite them to the, I was going to call it discord then, but to Buzz directly. And they can actually bring their own agents. So, you know, if your team or if people, you know, have AI agents and you'd like your agents to interact with their agents, then you can add them all to the same platform as well, which is really cool.

Also, there's a mobile app coming in the future and there's some experimental features you can plug on as well. So for example, we've got Pulse here, which is like a feed, an activity feed, kind of like a social media activity feed, but it's directly for your agents, not for you. So if we go back to the app here, we go to Pulse, we can see our agents. And if they create any notes or if they share anything, then we can see it inside this section as well.

I don't think it's particularly useful, but kind of fun to play with. And then over here, we have our different AI agents. So for example, we have Fizz, Honey and Bumble. And you can custom train these agents exactly how you want.

So you can direct message them as well, which is pretty useful as well. So we can direct message Fizz. And then also, if we go back, we can start using these agents and custom train them. So you can give custom instructions to each AI agent and you can choose which harness they are using as well.

Now, so far there seems to be three agents that you can add to Buzz. So you've got Claude, you've got Codex and you have Goose as well. And I'll come on to Goose in a second. So before we do that, let's have a look at the setup for the custom agent.

So you can change your profile picture. You can add the agent name, you can change the instructions. So let's say, for example, you wanted a SEO expert. You train the custom instructions here so that understands who it is and what it's going to be working on.

You can then also customize the agent harness you want to use for the agent. So for example, we've got Buzz, Claude, Codex, Goose, and we just need to install Goose as well. And then we can switch the model here. So if you want to use like a custom model, you can, or you can add Opus, Haiku, or Sonnet over here as well.

And then you can decide who can talk to this agent, which is really useful. So if you have like multiple different humans or members of your team that you've added to Slack, then you can decide who gets access to which agent. So if there's certain agents that have access maybe to your local files or something that you don't want other people to access to, you can just make them private like so. And then also you can decide how many different parallel conversations they can have at once.

So if you were speaking to an agent and other people were speaking to the agent, you'd need multiple different parallel conversations going on at the same time. So you can set those over here. The instance name pool, not really sure what that's about, but yeah, you can switch between these over here. Now you can add new agents and you can also change and organize your agents into teams.

So it's a really cool way to orchestrate all of your agents inside one place because you could have, for example, an SEO team, you can have a designer team, you can have a product team, and then you can set up your teams here. So this is how you set up the agents with custom instructions. This is how you set up the agent teams. Now over here, we can also change the agents that we've added to the platform.

So what does this mean? Well, basically we can have Codex, we've got BuzzAgent, we've got Claude, and then we can add Goose. So we can set up Goose over here and we can install Goose like so. Goose is an open source project as well that comes from Jack Dorsey's team.

And it's been trending for a while on GitHub. And what I think you could do with this is you get more customization on the APIs that you can plug in. So if you wanted to use a free API, for example, maybe like Ling 3.0 on OpenRooter, you could plug Ling 3.0 into Goose. And then I think you could use this for free as well, which means you don't even need to pay for ChatGBT or, for example, Claude or other subscriptions, which is really cool as well.

And then you can switch this on and off and you've got them all ready to go over here. So that's basically it. That's how to add teams, how to use the experimental features. You can also select which models you want to use over here.

So this seems to be able to use like local models, which is pretty cool. So it says, share this machine with your relay. When on, other members can run their agents here. So your other members can run their agents with these models, I think, which is pretty interesting.

I've not tested that out, but that could be quite cool. I also got a bunch of shortcuts you can add. And the way that I would recommend you get started with this is you just select the different workflows that you want. So if we run through some of the workflows, I've got this one for thumbnails, and then I've got an agent that can just create thumbnails, right?

And the great thing I like about that is like it's all siloed in one place. So we can come back to this whenever we need to generate thumbnails. And for each workflow that you work on day to day, you've got everything siloed away in one place. So you can come back to it whenever you need to.

So if we click on this, for example, we can say, we said, we said, like, you know, create a thumbnail for this image. And then it went off and created them, as you can see right here. The another example here, we said, right, OK, find some keywords for this. It looked for our search console, found us a bunch of keywords, and we've all got them ready to go over here.

And the same for blog creation. So, you know, we said create a blog about this keyword based on the SEO keywords over here. And then it started creating the blog and just published directly to WordPress. So it's very intuitive, easy to use.

You can plug in Claw, plug in Codex. You seem to be able to plug in Goose and you could use a free API with that if you wanted to. And then also you can add humans. They can add their own agents.

It's open source, free to use and very intuitive and easy to install because it comes in an app. So thanks so much for watching. If you want to get more training on this, we actually have a full Buzz masterclass inside the AI Profit Boardroom. Link in the comments and description or just go to the AIprofitboard.com.

Inside the community, you can ask questions, get help and support whenever you need to. And I personally answer these questions with a video tutorial every single day as well. Inside the classroom, you can get access to all of our best trainings and tutorials. So, for example, here you can see that we've got a six week beginner to expert course.

If you're a complete beginner, you can get our step by step guides and courses over here. And then we've got our new daily updates over here. So you can see that we add new daily updates based on what's just dropped and what's actually useful. And then we have our full agent OS system.

So if you like multi-agent orchestration like you've seen today, you can get our full system for that over here and a zip file to install it. Inside the calendar, you can jump a weekly coaching course, share your screen, ask questions, meet other members. And then inside the map, you can meet people in your local area. It's just a great place to connect with people, really get to know them.

Great place to connect with people, really positive, great to learn about AI automation. You can connect with me too. You can direct message me if you have any private questions as well. And if you're wondering, okay, what the reviews are like, you know, how, how has it been for everyone else?

So we've got over 204 pages of wins and testimonials and people learning and growing with AI automation. So loads of awesome people inside there. Lots of people are non-technical as well, so you don't need to be technical to get the most out of AI automation. We simplify as much as we can to help you.

And there's just lots of, you know, lots of wins, lots of positivity, lots of people learning and growing along the way. So if you want to get all of this is inside the AI Profit Boarding, link in the comments description, or just go to the AIProfitBoarding.com. Thanks for watching.