How to Create Your Own Custom GPT from Any Website with GPT Crawler

GPTs, or Generative Pre-trained Transformers, are powerful artificial intelligence models that can generate natural language texts on various topics and domains. However, not all GPTs are created equal. Some GPTs may be more knowledgeable and proficient in certain areas than others, depending on their training data and parameters.

Therefore, it would be beneficial to have a custom GPT that can understand and answer questions about a specific website that you are interested in. In this article, you will learn how to use a new open-source project called GPT Crawler to turn any site into a custom GPT with just a URL.

What is a Custom GPTs

Custom GPTs, or tailored AI chatbots, are groundbreaking tools for digital communication and automation. They’re specialized versions designed to cater to specific needs of websites and apps. These models are more than just utilities; they’re transformative, allowing businesses and developers to create personalized user experiences.

The key advantage of these custom GPTs lies in their adaptability. They can be incorporated into different platforms to provide tailored responses, guidance, or even coding support based on their specific training. This customization leads to a more effective and user-centric experience, addressing individual user queries directly.

What is GPT Crawler?

GPT Crawler is an open-source project that allows you to crawl a website and generate a knowledge file that you can use to create your own custom GPT from a URL. You can configure the crawler to match your needs, such as the URL to start from, the pattern to match for links to crawl, the selector to grab the inner text from.

Key Features

  • Customization: The ability to specify the URL allows for crawling any desired site, providing tailored content collection.
  • Depth of Knowledge: The crawler meticulously gathers information from designated pages, resulting in a comprehensive and detailed knowledge repository.
  • Flexibility: Adaptable to diverse content types, proficiently handling both publicly available and client-side rendered information.
  • User-Friendly Configuration: Simple configurations within the ‘config.ts’ file empower easy setting adjustments, including URL targets, pattern matching, and page limitations.
  • Efficiency: Utilizing a headless browser, the crawler operates efficiently and effectively captures dynamic web content, enhancing its overall performance.

How to use GPT Crawler to create a custom GPT

To create a custom GPT from a site, you need to first collect the text data from the site and use it as the knowledge base for the GPT. This can be done using a web crawler, which is a program that automatically browses and extracts data from the web.

GPT Crawler is a web crawler that is designed to crawl a site and output a JSON file that contains the title, URL, and text of each page. This file can then be uploaded to ChatGPT, a platform that allows you to create and share custom GPTs without any coding. Here are the steps to use GPT Crawler to create a custom GPT from any site:

Clone the repository and install the Dependencies

To get started, you need to clone the repository of GPT Crawler. You can do this by using this git clone command in your terminal. To obtain the GPT Crawler code, start by cloning the repository from GitHub. Use the command below to fetch the required files onto your local machine:

git clone

This will create a copy of the project on your local machine. Next, you need to install the dependencies of the project, which are the software packages that are required for the project to run. You can do this by using the npm install command in your terminal. This will install the packages that are listed in the package.json file of the project.

Configure the crawler with the site URL and other Parameters

Gpt Crawler

After cloning, you need to open the config.ts file in the code and provide your configuration. This file contains the parameters that control the behavior of the crawler. The most important parameter is the url, which is the starting point of the crawl. You need to enter the site URL that you want to crawl and create a custom GPT from.

Run the crawler and generate the output file

After configuring the crawler, you can run it by using the npm start command in your terminal. This will launch the crawler and start crawling the site. The crawler uses a headless browser, which means that it can render and interact with any web page, even those that are purely client-side rendered.

You can also customize the crawler to log into a site to crawl non-public information. As the crawler runs, it will print the progress and the results to the terminal. You can see the title, URL, and text of each page that is crawled. When the crawler finishes, it will save the output to the output file that you specified in the config file.

Integrate and Use Your Custom GPT

Gpt Crawler

After obtaining the output.json file from the GPT Crawler, you can directly upload it to ChatGPT. This involves creating a new GPT and configuring it with your file, thereby imparting it with the knowledge gleaned from your crawled pages. It’s a simple approach enabling interaction with your custom GPT through a user-friendly interface.

This method is suitable for those seeking swift deployment of a custom GPT for general use or testing purposes. For a more cohesive solution, particularly in product development, the OpenAI API provides a robust avenue. Here, you create a new assistant and upload your generated file.

This approach grants API access to your custom GPT, allowing seamless embedding into your products or services. It’s an optimal choice for delivering tailored assistance within your digital offerings, ensuring alignment with the specific knowledge pertinent to your product or content.

Frequently Asked Questions

What is GPT Crawler?

GPT Crawler is an open-source web crawler that can crawl any site and output a JSON file that contains the title, URL, and text of each page. This file can be used to create a custom GPT on ChatGPT.

What are the benefits of creating a custom GPT from a site?

Creating a custom GPT from a site has many benefits, such as leveraging the existing content and knowledge of the site, creating a unique and customized GPT.


In this article, I have shown you how to use GPT Crawler to create a custom GPT from any site with just a URL. You can use this tool to crawl any site and extract the text data from it. You can then upload this data to ChatGPT and create a custom GPT that can understand and answer questions about the site.

You can either share this GPT with others or integrate it as a custom assistant into your site or app. This is a fun and creative way to use the power of GPTs and create amazing experiences for your users. I hope you enjoyed this article and learned something new. If you have any questions or feedback, please let me know in the comments below.

#Create #Custom #GPT #Website #GPT #Crawler

Leave a Reply

Your email address will not be published. Required fields are marked *