Orca: New FREE AI Coding Agent is INSANE!

Today we're going to be looking at Orca, which is a new free open source. You can run parallel agents. You can basically have multiple terminals working together. You can also preview your projects. It runs with any coding agent and it's available on a desktop plus. You can run it on your mobile as well. So you can actually code from your mobile on multiple terminals locally, but using your mobile to control it, which is pretty awesome. So it's got some great features here. Mobile companion, you got parallel work trees, you got terminal splits, design mode as well, which is super nice. And then we also have the fact that you can basically run and load in GitHub projects and then preview those in a second. There's a bunch of other stuff like you can track your files inside there. I'll show you exactly how it works. Now, the other good thing about this is it's just a little simple app. So you can download Orca like so, and you can get the mobile app as well and then control and code from your mobile app. So once you've downloaded it, it's going to look something like this, as you can see. And then if we started a new project here, we can run it locally or we can actually run it on a remote Orca. So you can run it basically on a server as well. So you can run it basically on the VPS set. Now, when we're using this as well, you could have multiple different agents coding side by side on the same platform. If you want to create a new terminal window, let me show you how that works. So if we have a look here, we've got code running like so, but we could also close these and then I'll show you what it looks like step by step. So if we now add and right click here, split terminal right or split terminal down, we can choose between adding the terminal to right or adding the terminal down. So if we click on right here, we can now open up this terminal and we can just type in Hermes here. So we could have Hermes and Port Code coding simultaneously on the same project. The other good benefit about this is you can use free APIs with free open source projects for coding. And then, you know, let's say, for example, you want to delegate a task, but you're working on a project and you want like a free agent to work on that project because it's just like a little subtask. You can do that over here. So for example, we've got Hermes agent that is free and we have lean 3.0 flash plugged in. That is also free. And so we can use Hermes agent for free here. And then we've got different terminal windows working on the same projects at the same time. Now, also, if we go to our settings over here, we can switch between and add two different skills. So you'll see orchestration and you can basically set up orchestration. Definitely recommend that. That basically enables your agents to hand off context and coordinate work through Walker. And then you can also allow computer use. So you can allow computer use as skill. I'll show you how to use those in a second. And then over here, we can switch between the default agent that we want to use. So if you've got multiple agents installed locally, then you can switch between them and you can add them all in. So, for example, we've got Quod, Quodation Teams, Codex, OpenCode, Gemini, Antigravity, Kimi, Hermes and OpenCore. If we scroll down here, we can disable or enable them and we can also decide which one we want to be the default. We can also code with voice as well. If you want to enable voice dictation mode. I already use Wittable Flow for that, so I'm not going to install it, but you can use that. And then we've also got different integrations here. So we can add, for example, Git and plug that in too. And then we have the mobile option here. So this is how you can pair a phone. Basically, what you can do is just quickly generate the QR code. Then you scan it in Orca Mobile on your phone and then you can pair your device with this setup. And that's pretty good as well because right now I would say the Quod Remote is not that great when I've tested it out. Whereas this might be a better option. You can control and code locally with this setup. Now, you can also have multiple different projects. So for example, we could set up a new project over here and we can switch between smart, dev, branch or name. That means we can actually find an existing project and then branch. And we can rename it like so. So if you right click and click update, then we can rename this. So we can put, okay, code code over there. Now what you can see is we have a local host for this project, right? So let's say you were calling our website. We've got the local URL here. We can actually open up a new browser window. So if you click the plus sign at the top here, then you can select new browser tab. Then you can select the URL of the project, which is this one that we're working on right now. And from here, not only can you preview it, which is pretty cool because then you have the terminal and the live preview. But also you can draw on the screenshot and give notes or annotate it to your AI agent. So let me show you what I mean. So for example, we can grab a page element like this one, copy that, grab that element, go back into Claude here. And we can say, okay, make that more interesting, beautiful. So we can select any of the elements on the page and ask it to improve it, which is really good. Because if you're not a coder or you're not a developer like I'm non-technical, then you can easily use this tool like this. And it doesn't take a lot. You can also import your browser data here if you want to, if you just want to use it like Chrome. And we can annotate the page as well to give feedback to the agent. So if we were having a look at this one, for example, we could say, okay, improve the headline. So it's more direct response style copywriting and more interesting. And you can either ask a question or make a change. So it depends on what you want to do here. So you want to change it up? You can. If you want to add a question, you can. So from here, we'll click on change. That's added the annotation. And then once you've finished all of your annotations like so, you can actually send that to Claude or to a different agent team. So it could be, for example, Claude agent teams, could be Codex, OpenCode, whichever you prefer. So we can send the notes to the live agent over here, which we've just done. And you can see our feedback down there. Now, if we add another tab here, we can actually set up like a markdown file or we can emulate a mobile. So let's say, for example, you're working on an app. You could preview the app and emulate it over here. And then we can see the app we're working on. So if we go inside the tab here and just say, create a simple mobile app for habit tracking. And then we'll add a new tab and we'll call this new mobile emulator. We'll close this one. And now we've got the code on the left-hand side and the emulator on the right-hand side. You just need to make sure that you have an emulator device. So for example, like Xcode or for example, Android virtual device in Android Studio, if you want to do that. Then if you need to close anything, you can close it like. You can also easily fork for conversation too. So if you click on, if you right-click, then click fork agent session, you can just create a new version like that. And then you've got multiple different conversations with terminal windows based on what you're trying to do. Now, if we refresh the page, you can see that it's beginning to look a lot more interesting. It's added the new logo. It's just a better looking page. There's also a bunch of experimental features. If you're familiar with codecs, for example, you can add a pad like so. And this will pop up so that if you're, for example, in your browser like this, you can see what it's working on. Again, these are new features. So they're still in beta. Then we have agent view and agent dashboard as well, plus a chat UI too. So if we go back to the dashboard now, you can see that we've got agent dashboard and the agent section as well. And then we can see who's working, who needs help and approval. So it's quite easy to see, okay, what's going on. And then we can click into that and X off that and that's just working in the background. So basically this avoids, for example, if an agent needs your permission or something like that, this avoids it just being idle in the background, waiting for your approval. So you can quickly see what's going on. Now, there's also this automation section. So all this can do is basically scheduling tasks, kind of similar to Hermes. In fact, if you click on add new, we can choose which agent we want to work in this. So, for example, if we type in Hermes, we can actually have Hermes working inside the project. And then we just set up a schedule for it to work. So we've got weekdays here, 12 hours, et cetera. You might also say, why don't you use, for example, Hermes desktop, but that's not a multi-agent orchestration platform. This really is. So this is an app where you can orchestrate multiple different agents. You can even switch over to the Hermes tab over here inside automations. So you can just select Hermes instead. And then you can say whatever schedule tasks you want to use. You can also click on weekday, sorry, use template. And we've got multiple different templates that we could go through as well. So if you wanted to schedule a recurring task, you do every single week. It's quite easy to do it over here. And also you can manage that from your phone. The other thing that's interesting is that you can have multiple different projects. So it's kind of more productive because if you're waiting for agents to code out like we are, then you can have multiple different projects that you're working on. It could be like one's a mobile app, one's a website, one's, for example, a game. And you're just coding across multiple different projects at the same time in parallel. So it's much easier to use and you don't have to sit there waiting for every agent to get approval or to finish coding their task.