News

GitHub Expands Copilot Enterprise Search in Visual Studio and VS Code

GitHub supercharged search for its Copilot Enterprise AI assistant in both Microsoft's Visual Studio IDE and Visual Studio Code so developers can now get results from well beyond local codebases, including the internet.

GitHub Copilot Enterprise shipped early this year, offering a new way for teams to use the AI-powered code completion tool in a customized experience providing enterprise-grade security, safety and privacy along with nuts-and-bolts coding help for $39 per month, about twice the price of the regular subscription.

GitHub Copilot Enterprise
[Click on image for larger view.] GitHub Copilot Enterprise (source: GitHub).

At the time, GitHub noted the tool can go beyond organizational boundaries and knowledge bases to bring in internet information to aid its work, thanks to beta integration of Bing search directly into Copilot Chat, the company's popular Copilot companion tool that uses advanced natural language processing (NLP) to understand human speech and provide code suggestions in real time.

GitHub Copilot Plans
[Click on image for larger view.] GitHub Copilot Plans (source: GitHub).

This month, that functionality advanced to preview status, with GitHub announcing:

  • GitHub Copilot Enterprise users in Visual Studio Code can now get answers from Copilot Chat enriched with context from Copilot knowledge bases and Bing search results.
  • GitHub Copilot Enterprise subscribers in Visual Studio can now use Copilot Chat to get answers enriched with context from their entire repository (not just open tabs!) and Bing search results.

Here, the term "knowledge bases" has a specific Copilot meaning, as it refers to a compilation of Markdown documentation from multiple repositories, which organization members can use as context for Copilot Chat to help the tool enhance code suggestions and responses to user queries. Documentation on creating a knowledge base is available here.

Visual Studio Code
Now, VS Code users, armed with latest release of the GitHub Copilot Chat extension for VS Code, can just type @github #kb from any Copilot Chat conversation and select a knowledge base from the provided list to enter a question.

"Copilot will respond, using the Markdown documentation in your knowledge base as context for its answer," GitHub said. "This allows Copilot Enterprise users to combine search results and internal documentation with editor context using existing chat variables like #file and #selection."

To get answers enriched with Bing search results (which must be enabled by an administrator), VS Code devs can start a message with ,@github, upon which Copilot will intelligently decide when to use Bing. Devs can also ask the tool to run a search by adding #web to a query. So an example query could be: @github What is the latest LTS version of Node.js? #web.

Using Both @github and #web in One Query
[Click on image for larger view.] Using Both @github and #web in One Query (source: GitHub).

Visual Studio 2022
For the IDE, users must install Visual Studio 17.11 Preview 2 at least to use the functionality.

"Copilot Chat can now answer questions with understanding of your full repository, not just the tabs you have open," GitHub said.

"Index your repository on GitHub.com, and then ask a question mentioning @github. You can ask questions like @github Where is device detection implemented?"

As far as bringing in internet info, Visual Studio developers can type @github to have Copilot intelligently decide when to use Bing, just like with VS Code. Another similarity: Bing search must be enabled by an administrator. Documentation on enabling GitHub Copilot Enterprise features is available here.

About the Author

David Ramel is an editor and writer for Converge360.

comments powered by Disqus

Featured

  • Logistic Regression with Batch SGD Training and Weight Decay Using C#

    Dr. James McCaffrey from Microsoft Research presents a complete end-to-end program that explains how to perform binary classification (predicting a variable with two possible discrete values) using logistic regression, where the prediction model is trained using batch stochastic gradient descent with weight decay.

  • Dev Asks, and 7 Years Later Python in VS Code Delivers Django Unit Test Support

    "We are excited to announce support for one of our most requested features: you can now discover and run Django unit tests through the Test Explorer!"

  • OData Finally Ditches Old .NET Framework

    "The most disruptive change we are making in this release is dropping support for .NET Framework."

  • .NET MAUI, ASP.NET Core Polished in First Release Candidate for .NET 9

    Microsoft shipped the first release candidate for .NET 9, which is nearing feature completeness and production readiness in advance of its November debut.

Subscribe on YouTube