
In this video, I am excited to introduce FireGeo, an AI-powered brand visibility platform designed to kickstart your SaaS application. FireGeo leverages advanced tools such as Firecrawl for website analysis, provides competitor tracking, automated prompt generation, and detailed visibility metrics across multiple AI platforms like ChatGPT, Anthropic, and Perplexity. The platform also includes user authentication, subscription management via Stripe, and a credit system powered by Autumn. You'll find a detailed README and have options for both automated and manual setups to get started. If you find this resource useful, please like, share, and subscribe! Star the repo! https://github.com/mendableai/firegeo 00:00 Introduction to FireGeo: AI Brand Visibility Platform 00:09 Brand Monitor Page: Adding and Analyzing Websites 00:41 Generating Relevant Prompts and Custom Queries 02:30 Comparison Matrix: Brand Mentions Across Platforms 03:46 Provider Rankings and Share of Voice 05:54 Authentication and Basic Functionality 06:27 Subscription Plans and Stripe Integration 07:26 Autumn Service for Billing and Credits 08:42 Setting Up the Application: Automated and Manual Steps 09:47 Conclusion and Call to Action
--- type: transcript date: 2025-07-16 youtube_id: hEHBcPsskZs --- # Transcript: Introducing FireGeo: The Open Source AI Brand Visibility SaaS Starter Kit In this video, I'm excited to open source Fire Geo, which is an AI brand visibility platform, and it's equipped with everything that you need to get started with building a SAS application within the platform. The core functionality of how this works is if you go over to the brand monitor page, what you're going to be able to do is you can put in a website and then as soon as we submit it, we are going to be leveraging fiber crawl to get the context of what that website is. From there, we can automatically identify customers. Additionally, within here, we can put in other competitors if we have any in mind. So, say if I want to add in Firebase, for instance, you can also remove some if you don't want to have particular ones within here. You can add as many as you'd like within here. And then, as soon as you're ready, you can go to perform the analysis. Right off the bat, what it will do is it will generate relevant prompts based on what the website is. In this case, since I leveraged Verscell, we do see that there are a number of different prompts like most popular hosting platform today or top hosting platform for startups. And the concept around these prompts is you can imagine leveraging something like chat GPT and you're looking for a hosting platform or you're looking for something to scrape a website. The idea with this is ideally you would have your brand within the visibility of these different AI platforms. So effectively the goal with the platform is to help you identify what different queries actually render, whether it's you or your different competitors. By the end of this, you'll be able to see the visibility of the types of queries that you rank in or don't rank in. And it can start to give you a potential idea around content strategy, maybe different blog posts to begin to rank within some of these things. Or maybe you see that some of these results are generated from things like technical tutorials or what have you. Additionally, what you can do within here is you can also add in your own custom props. So if I say best place for a Nex.js app and I add in that prompt. Now what I can do with this is I can go ahead and I can submit this and it's going to go through and one by one is going to send in all of those different prompts to each of the respective providers whether it's perplexity chat GPT anthropic or what have you. Now additionally what's built in within here is as you perform more analysis you will be able to revisit these. So we do have all of the logic set up where as soon as this analysis is complete, it will log within the sidebar here where you can go and you can revisit past previous queries of different websites. So in this case, I did use Versel previously and that's why we have it there. But if I had in other websites, you can see them all stored here. This information on the sidebar here is going to be persistent to whatever user is logged in. Now I'll go ahead and close this out. Now just to touch on the comparison matrix and how this works. Effectively what this is, how many times is your brand mentioned across all of those different prompts. For instance, for Verscell, it is within 100% of the responses from Anthropic, 20% of those within chat GPT, and 80% of those within Perplexity. And within here, you can also see the ratio of the number of times that the relevant competitors were also mentioned within these other platforms. So now, if I hop over to prompts and responses, here are all of the different prompts. Here are all of the different platforms. And what you'll notice within here is we do have this pill that indicates that Verscell was mentioned in every single prompt at least once across the different providers. Now, additionally, at a bird's eye view here, we can see the different platforms where Verscell was mentioned. So, say for instance, if I open up the best hosting platform in 2025, I do see that it wasn't mentioned within OpenAI, I see things like AWS, Azure, GCP, as well as Digital Ocean and Blue Host. But if I go on to anthropics response for the same query, I do see that Verscell is in fact listed here. Now, additionally within here, you can plug in things like if you want to look up particular competitors. Say if I'm interested in something like Heroku, I can see all of the different mentions every time that particular competitors or particular keywords were mentioned within this. Now, next up within this is provider rankings. So, what this will show you is across all of the different brands, how consistent is your brand showing up in those different prompts that we put in. Like you saw on the first matrix there within Anthropic, I have 100% visibility. We also have the share of the voice. So this is going to be the percentage across all of the different mentioned brands in terms of how many times your brand comes up relative to some of the other brands. For instance, I do see on chat shebt that Heroku actually does have a higher share of the voice as compared to something like Verscell for those particular prompts. And if I hop over to Perplexi, I do see again similar to Anthropic that Verscell does perform quite well, but it doesn't have quite as high as a share of the voice as we see on something like Anthropic. All of these different metrics begin to give you an idea on which competitors do actually perform well in terms of what they're doing for their content strategy to actually get ranked within the responses of things like chat GPT. And the thing with geo or generative engine optimization is this is increasingly going to be the term that you hear analogous with something like SEO. While SEO is still very important, increasingly so over the coming years, it is going to be very important to see how these different platforms actually include or don't include your brand. The idea with this is to start to give you an idea on how you can actually get insight with your brand on what it's actually doing. And then additionally within here, we do just have another representation of the data here for the visibility score. So within here I can see the overall score across the board all of the different prompts was 66.7% which is actually really good. So Burcell is doing very well with whatever they're doing in terms of content and being able to rank for all of the different responses even for some of those more general things like just getting up there to be included with things like Azure or GCP. Now mind you I did include a prompt that was Nex.js JS specific. So that does bias the results a little bit there. But just the fact that it does rank with some of those generic results definitely is impressive. I'd encourage you try this out with your own brand. Play around plugging in some of your competitors as well as some different prompts and see what types of results that you get. So now just to go through some of the functionality. It is equipped with authentication. It is leveraging better off for the framework as well as Postgress. So you're going to be able to use Subabase Neon whatever Postgress provider that you would like to use. It does also have some basic functionality in terms of things whether you want to set your display name or include your phone number or bio. This is more just to show you how this is actually going to persist within the database. You will be able to see all of that information logged within whether it's Superbase or Neon or whatever you're leveraging. Now, additionally, within here, we do also have the ability to select between the different plans that we want to use. If I go over to plans, what we'll see within here is we have the free plan and then we have the pro plan. What you'll see within here as soon as I select the free plan, it will go ahead and provision that amount of credits within the system. Now, additionally, it does have the functionality where if you do have a pro plan and you do want to specify a certain number of credits, you can go ahead and upgrade just like this. And what it will do is it will load the checkout. And all of this is set up to function with Stripe. So, as soon as we put in the information, so if you're testing this out, you can go 424242, etc. put in an expiration date within the future and then just put in some madeup information and you will be able to test this out within the Stripe test environment. So, I'll just go ahead and subscribe here. And as soon as I subscribe, what we're going to see is it's going to redirect us back to our application and we're going to be able to see that we're now on that premium tier. So now within here within our dashboard, I do see we have this pro tier. I could have indicated this with maybe having a th000 credits on the pro tier or something, but what's really nice with how this is set up is we're leveraging a service called Autumn. And Autumn is great. It allows you to have, I think it's $3,000 MR before they actually even begin to charge you. And it has all of the nice functionality built in. So, you're going to be able to add in products, usage based pricing just like I have within this application. and all of those different pieces that you would have otherwise had to wire up yourself, you're going to be able to leverage with the nice capability that is built in to Autumn. Here I see that I specified to downgrade to the free plan once my pro plan is done. And we do see that next month I'm not going to be build and it's going to go down to that free tier. All of that billing portion you really don't need to worry about with Autumn, which is really great in terms of what they built out. Now, additionally, what I did include within here is a basic chat page. And the purpose of this is just to really show you the credit system working with Autumn. So if I say something like hello, we'll be able to see it deduct from the Autumn backend. And what you can do within Autumn is you can take a look at each respective user. And you can see all of the different users that your application has as well as how many credits that they have. And if you want to bump up different credits, let's say you have a support request and you want to give a user more credits for whatever reason, you can go ahead and do all of that really easily within Autumn. All that you need to get started is you can go ahead and you can pull down this repo and I did also include a very detailed readme in terms of how to set up all of this and effectively there are two ways. So you can set this up with an automatic process where it will go through and so long as you have all of the relevant API keys as they're specified within the readme. You can go ahead and run the setup script, it will go through and set up the products on autumn and set up all of the relevant tables within Postgress and basically all of the different component pieces to get started with everything. It will all happen from this automated setup script. So once you've done that, you can go ahead and mpm rundev to have your development server or alternatively you should be able to just deploy the template. So there should be no build issues or anything like that. Now additionally, if you do run into any errors with the automatic setup script, what you can do is there is also the manual setup steps within the readme. So this will just walk you through step by step effectively what the automatic script is doing, but it will give you a little bit more control to actually see the visibility. If you do run into any issues, you can just go ahead and tweak and realize, okay, oh, maybe I forgot this API key or I added an extra character there or what have you. So, otherwise, that's pretty much it for this video. This definitely took a ton of work, so I definitely appreciate a star on the GitHub repo if you don't mind. But if you like these types of templates, let me know in the comments below. The best way to support this type of work is start the repo, like the video, share this, and engage with it. And if you have any ideas in terms of what you'd like to see for these types of full stack templates, let me know within the comments below. But if you found this video useful, please comment, share, and subscribe. Otherwise, until the next
Weekly deep dives on AI agents, coding tools, and building with LLMs - delivered to your inbox.
Free forever. No spam.
Subscribe FreeNew tutorials, open-source projects, and deep dives on coding agents - delivered weekly.
Technical content at the intersection of AI and development. Building with AI agents, Claude Code, and modern dev tools - then showing you exactly how it works.