Transcription
A while back, we had covered Augment Code, a revolutionary agentic AI IDE with one of the most impressive context engines we've ever seen. Built directly within your development environment, Augment Code stood out as a powerful AI software development platform where autonomous software agents operated seamlessly within your IDE and within the cloud. And just today, they've taken things to the next level with the launch of Auggie CLI, bringing their agentic development powers straight to the command line.
Auggie is now available to everyone, equipped with deep codebased intelligence from Augment's industry-leading context engine. Now, let's just pause for a second. I know what you may be thinking right now. Seriously, we've showcased so many different CLI tools. What is making this one different from something like Cloud Code or Google's Gemini CLI?
Well, Auggie CLI sets itself apart from all of these different CLI tools because it has an advanced context engine, which is enabling it for far deeper understanding of large code bases. This means you're not just coding with AI, you're elevating your entire development environment, which is going to reduce hallucination, save a lot of tokens, which is going to make it super cheaper for you. And there's going to be a lot of added benefits by deploying AI agents that have the same functionality with something like Cloud Code or Gemini CLI.
In short, Auggie CLI has the ability to automate workflows using custom commands like you would with Cloud Code. You can access IDE-grade features within the terminal, including task list prompt enhancer as well as a model picker. You can even define tool permissions for precise control, perfect for CI/CD and automated systems. And one thing that I really like is that you also have ready-to-go GitHub actions for AI-powered code reviews as well as PR descriptions. So you have basically everything integrated within the CLI. You can even integrate everywhere with Unix-like pipe support, Node.js, from local terminals to GitHub actions and even AWS Lambda. In short, this is a CLI that's going to transform your terminal into a powerful AI-driven coding agent that's capable of editing files, running commands, and so much more.
Now, to get started, first things first, you need to make sure you have the prerequisites fulfilled. Make sure you have Node.js 22 and above. Once you have that installed, you can head over to the CLI page, which I'll leave a link to in the description below. You can simply go ahead and install Auggie with a simple single-line command with npm. So, just simply go ahead and open up your command prompt. Once you have it opened up, you can simply go ahead and paste in this command that we have over here within the terminal and you can click enter. This will start installing all the packages necessary to have Auggie functional.
Now, to start it up, it's really simple. So just go over to your terminal and type in `Auggie` and this will load it up. And this is where you can simply go ahead and either log in with an Augment Code account or you can create one for free. After signing in or creating an account, what you want to do first is enable indexing towards your overall codebase or your workspace. This is pretty important because if you want to use the overall context engine that Augment provides, you want to make sure that you set the correct path while working with your overall codebase.
Also, just in FYI, their co-founders, Guy and Iger, are doing an AMA on Reddit. So, if you have any questions about AI dev tools or just building products, you can drop them over here, which I'll leave a link to in the description below. Once all that's complete, this is where you're going to be introduced to the main command line of Auggie. You can simply go ahead first and type in `/`, which will showcase all the different commands like automations, using different agents, the editor, feedback, GitHub integrations, so many other things. And if you want to get a better idea of all the commands and see what they're actually capable of doing, you can go over and type in `help`. And this will give you example prompts, different commands, and explain what each of them do.
Now, first things first, you want to set the model. And if you type in `/model`, you have the ability to select from two different models: the GPT-5 model or the Claude 4. I recommend you use this one. And you can click enter. Then within your workspace, you can type in any prompt like "Create me a task management app with many features like a drag-and-drop note-taking function, a calendar, and a couple of other things." But what you can also do is if you press Ctrl+P, this is where you can send it into the CLI to enhance your prompt. This is a feature that Auggie provides which will overall make your prompt more detailed so that the model can easily follow along with whatever you are trying to create. So in this case, if we approve this prompt that it has generated, we can send it in and right away it's going to work on deploying multiple agents to accomplish this task.
What's nice about Auggie CLI is that you can actually see diff changes directly within the terminal. So you get a good visualization of what is being changed or edited. And if you can see over here, it looks like it has finished generating all the code. And you can see how well it is in terms of easily starting up this project. It created the base structure and we even have an authentication system that it developed. After logging in, you're going to be then sent over to the main dashboard of our task management app, which looks pretty nice. You have the main dashboard and it has listed out all the features that we had asked for, like a task list where you can add in different tasks with different priorities by simply clicking on this button. You also have a calendar which we had also requested, and it has also generated and displayed all the different action items that would be created from the task list. There's also categories. You have different categories of different tasks that you can organize within. There's even a settings page. Overall, it did pretty decent in terms of this generation, and it did it pretty quickly.
And remember guys, every time you are working on something new with Auggie, it's going to be able to index every single time. And this way, it's going to be able to catch up and have the latest context of your overall codebase. I'm simply going to go ahead and ask it to add more features. So, let's see what it's actually capable of adding further. And just from a single prompt, you can see that it has added a couple of other new pages like a time tracker. You have templates that you can access like a daily stand-up meeting template that will help you prepare for and attend daily stand-up meetings. And there's also categories. So overall, it did a decent job adding new features. And this is where you can easily prototype and build various sorts of applications quite rapidly.
Also, there is a task manager. So this way you can actually run asynchronously, meaning you can develop and deploy different agents that can run simultaneously to accomplish multiple tasks. And since this is one of the best context engines that are available for a CLI tool, I highly recommend that you use this for debugging because in this case, it's going to be able to thoroughly access all of your code files and it's going to be able to analyze it with better context understanding than other CLI tools would be able to do. So in this case, I simply just asked it to find errors if there are any. And right away you can see that it is working on reading multiple files and understanding our codebase and trying to see if there is any errors that it can find. Now it looks like it is thoroughly showing all the different diff edits that it has made and it is trying its best to find and resolve all of these errors within the overall codebase.
At the start, I did mention that there are different integrations where you can connect MCPs, but you also have a direct integration where you can automate different sorts of things like code reviews, issue triages, automate on-call, as well as have an exception management through GitHub. And it's super easy to set up. Once you have Auggie installed, all you got to do is first log in with Augment. And then you just need to output your token. And from there you can pass the token to Augment using your GitHub authentication. So I've provided my GitHub token and I have authenticated it. Now what I can do is I can say something like to Auggie to open up a GitHub issue for the bug in the user login functionality. I can simply go ahead and send this in and it's going to be able to process it and make the changes onto the GitHub repository that I have. And you can see right now that it has got my repository information that I had provided. And you can see it is the DeepSseek webcrawler that I have, which is a test repo. So now it's going to be able to upload the issue onto that repo. And there we go. We now have this login functionality bug within the issue tab of the Deepseek webcrawler repo. If you click on it, you can see it has opened up and created this new issue.
If you like this video and would love to support the channel, you can consider donating to my channel through the Super Thanks option below. Or you can consider joining our private Discord where you can access multiple subscriptions to different AI tools for free on a monthly basis, plus daily AI news and exclusive content, plus a lot more.
Overall, there's so much more to this. This is just the tip of the iceberg. There's so many other functions and areas that we haven't even covered like the multi-models that you can operate simultaneously. You have the user repo rules that you can set `/commands` and so much more. I highly recommend that you take a look at this because this is truly an amazing new tool that is better than most CLIs due to the context engine that is built within the CLI tool itself. I'll leave all these links in the description below so that you can easily get started, especially this AMA, which I definitely recommend that you attend. Make sure you go ahead and join the newsletter if you haven't already. Subscribe to the second channel, join the Discord, follow me on Twitter, and lastly, make sure you guys subscribe to the YouTube channel, turn on the notification bell, like this video, and please take a look at our previous videos because there is a lot of content that you'll truly benefit from. But with that thought, guys, thank you guys so much for watching. Have an amazing day. Spread positivity, and I'll see you guys fairly shortly. He suffers.