The Deep Research agent, which was announced last week, will allow users to connect their GitHub repository to ChatGPT. According to OpenAI, the new feature will enable the AI chatbot to generate detailed reports based on the user’s and other users’ repositories. At the moment, the San Francisco-based AI company is only implementing the feature for the platform’s paid subscribers. Interestingly, it can take up to five minutes for a newly created or edited repository to start appearing in the Deep Research GitHub connector.
ChatGPT’s new deep research integration could be your next indispensable tool if you deal with code from several GitHub projects. This functionality, which is now in beta and being made available to ChatGPT Plus, Pro, and Team members, gives your GitHub content real-time reasoning along with the strength of live data access.
Deep research may retrieve code, README files, and documentation from your repositories once it is connected. For example, “Where do I handle file uploads?” and it will, in a matter of seconds, search your repos, scan the code, and cite the precise bits that are important.
The announcement of the new functionality was made by OpenAI Developers’ official handle on X, formerly known as Twitter. Team users can access ChatGPT’s Deep Research with GitHub from anywhere in the world. The feature will also be available to all Plus and Pro subscribers, with the exception of those in the UK, Switzerland, and the EEA.
A video demonstration of the feature was also uploaded by the account. The qualified users will next need to pose a question to Deep Research. They will then see a new GitHub connection button, which they can tap to grant the chatbot access to their repository and any other public repository owned by a third party.
According to OpenAI, “ask a topic and the deep research agent would read and search the source code and PRs in the repository, returning a detailed report with citations.”
And if you are concerned about what is disclosed, then you should not be. You are always in charge of which repositories are available, and ChatGPT only sends search queries to GitHub in response to your prompts. Your data isn’t automatically used to train models if you’re using ChatGPT Team or another business solution.
It may take up to five minutes for certain repositories to show up. If yours isn’t visible, it might be private, recently made, admin-blocked, or not yet indexed by GitHub’s search engine. This can be fixed by manually starting indexing or by adjusting your GitHub settings.
Team users can access ChatGPT deep research with GitHub from anywhere in the world. With the exception of users in the UK, Switzerland, and the EEA, the GitHub deep research connector is currently completely available to Plus and Pro users.
An announcement on enterprise user access will be made later.
Deep research can now be connected to your GitHub repositories. Currently in beta, this function is accessible to ChatGPT Team users worldwide as well as Plus & Pro users everywhere save Europe. Please be aware that third-party services like GitHub have their own terms and conditions and are connected apps.
The AI company noted in its help sites that GitHub’s terms and conditions will be applicable and that the tool is presently in beta. ChatGPT may retrieve real-time data from repositories, such as code, Readme files, and other documents, when Deep Research is connected to GitHub. In order to offer insights regarding the content, the chatbot can also analyse the data.
OpenAI also notes that it does not store data from its business offerings, such as Team, Enterprise, Edu, and its application programming interface (API), but if users of ChatGPT Plus and Pro have left the “Improve the model for everyone” setting enabled, the company will use the GitHub data to train its models; if the setting is off, the data will not be stored.
We use cookies to ensure that we give you the best experience on our website. If you continue to use this site we will assume that you are happy with it.
