GitHub Copilot Chat is a chat interface that allows developers to ask and receive answers to coding-related questions directly within a supported IDE. It is currently in open beta and available for all GitHub Copilot individual users across Visual Studio and VS Code.
GitHub Copilot Chat was introduced as a public beta as an AI-powered software development extension for all GitHub Copilot for Business users. Now, it is enabled for all Copilot for Individual users for free, supported in both Visual Studio and Visual Studio Code editors. The extension leverages advanced natural language processing (NLP) to offer assistance and answers in natural language directly within supported editors. It eliminates the need for developers to consult documentation or search online forums.
Shuyin Zhao, VP of product for GitHub Copilot, writes in a blog post:
By reducing the need for context switching, it streamlines the development process, which helps developers maintain their focus and momentum.
GitHub Copilot Chat also enables individual contributors to propose security patches, thus enhancing the overall security of open-source projects. In addition, other features of Copilot Chat detailed by GitHub include:
- Real-time, tailored suggestions for best practices, tips, and solutions
- The breakdown of complex concepts with explanations of code snippets
- Recommendations for remediation and helping to reduce the number of vulnerabilities found during security scans
- Identifying issues during debugging with tips, explanations, and alternative approaches
GitHub Copilot Chat in Animated Action (source: GitHub blog post)
With the availability to more users and what Copilot Chat offers, there’s still a demand.
A developer, Haseeb Heaven, a developer, commented in a tweet:
Bring features like Auto debug, knowledge of codebase, and please start using GPT-4, for God's sake, People already switched to @cursor_ai for more advanced features.
And, SMB Punt, tweeting:
Only Visual Studio? When JetBrains?
With Copilot Chat’s availability, GitHub conducted research on the extension, and the conclusion was that "the quality of the code authored and reviewed was better across the board with GitHub Copilot Chat enabled, even though none of the developers had used the feature before."
Research results from GitHub Copilot Chat (Source: GitHub blog post)
Yet, Rafa Darader questions the research as he tweets:
Hey, I love copilot, but this study hardly talks about quality. The metrics outlined are either how developers feel with the addition of copilot to their workflows.
Can quality be measured by only looking at the parts involved in the production of the thing?
Lastly, to leverage Copilot Chat, users must have an active GitHub Copilot subscription, and VS Code users must have the latest addition and be signed into VS Code with the same GitHub ID that has access to GitHub Copilot.
