Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

copilot-enterprise

Subscribe to all “copilot-enterprise” posts via RSS or follow GitHub Changelog on Twitter to stay updated on everything we ship.

→ ~ cd github-changelog
→ ~/github-changelog|main git log main
showing all changes successfully

Copilot Chat in GitHub.com is now contextually aware of GitHub Advanced Security alerts

You can now use Copilot Chat in GitHub.com to search across GitHub to find and learn more about GitHub Advanced Security Alerts from code scanning, secret scanning, and Dependabot. This change helps you to better understand and seamlessly fix security alerts in your pull request. ✨

Try it yourself by asking questions like:
– How would I fix this alert?
– How many alerts do I have on this PR?
– What class is this code scanning alert referencing?
– What library is affected by this Dependabot alert?
– What security alerts do I have in this repository?

Learn more about asking questions in Copilot Chat on GitHub.com or about GitHub Advanced Security.

See more

Copilot Chat in GitHub.com now can search across GitHub entities

Image

With this change, you can now use natural language within Copilot Chat in GitHub.com to search across GitHub to find commits, issues, pull requests, repositories, and topics.

Try it yourself:
– What are the most recent issues assigned to me?
– What repos are related to [insert topic]?
– What is the most recent PR from @user?

We’ve also made some changes under the hood to make Copilot more efficient with how it stores conversation histories. This means that Copilot can now remember more of the history of your conversation which should result in more informed and reliable responses ✨.

Join the discussion within GitHub Community.

See more

Custom models for GitHub Copilot are now in Limited Public Beta

Custom models for GitHub Copilot are now available in Limited Public Beta for Copilot Enterprise. This new capability lets you fine-tune Copilot to better understand and align with your organization’s unique coding practices, improving the relevance and accuracy of code suggestions across your projects.

What are custom models?

Custom models are large language models (LLMs) that have been fine-tuned using your organization’s codebases. By training a model on your proprietary libraries, specialized languages, and internal coding patterns, Copilot delivers code suggestions that are more context-aware and tailored to your organization’s needs.

During this beta, you can create a custom model using your GitHub repositories. Optionally, you may also enable the collection of code snippets and telemetry from developers’ Copilot prompts and responses to further fine-tune the model. This process closely aligns Copilot’s suggestions with your coding practices, making them more relevant and accurate. As a result, your development teams will spend less time on code reviews, debugging, and manual code adjustments, ultimately boosting team productivity and ensuring more consistent code quality.

Custom-Model-Training-Config

Importantly, your data remains entirely yours. It is never used to train another customer’s model, and your custom model is kept private, ensuring full control, security, and privacy.

When to Use Custom Models

Custom models enable you to make Copilot’s suggestions more relevant to your specific needs, which can lead to higher acceptance rates of the code suggested by Copilot among your developers. Consider using custom models in the following scenarios:

  • Enhance Library and API Usage: When your organization relies heavily on custom libraries or APIs that aren’t well-represented in public datasets, a custom model can prioritize these in its suggestions, making it easier for your developers to follow internal standards.

  • Improve Support for Specialized Languages: If your team works with less common or proprietary languages, custom models can make Copilot much more effective. Fine-tuning helps Copilot understand these languages better, reducing friction and improving productivity.

  • Adapt to Evolving Codebases: As your codebase changes, you have full control over when and how often to retrain your custom model. By regularly retraining, you can ensure that Copilot keeps up with the latest coding patterns, so it continues to provide relevant and accurate suggestions.

How to Get Started

  1. Sign Up for the Beta:
    Sign up here to participate in the Limited Public Beta and make sure your organization is on the Copilot Enterprise plan.

  2. Prepare Your Repositories:
    Choose the repositories that best reflect your organization’s coding standards. Include those with proprietary libraries, specialized languages, or key internal frameworks to get the most out of fine-tuning. If your enterprise has multiple GitHub organizations, note that only one organization and its repositories can be used for training during this beta.

  3. Enable Telemetry Collection:
    To further customize your model, consider enabling the collection of code snippets and telemetry related to developers’ prompts and Copilot’s suggestions. This data will be securely collected and used for additional fine-tuning, improving the accuracy and relevance of Copilot’s output for your team. Your data will only be used to enhance your custom model and will not be shared with others. For more details about our data-handling practices, please visit the Trust & Security Center or review GitHub’s data protection agreement.

  4. Training and Usage:
    After setup, your custom model will be trained using the selected repositories. Once it’s ready, your developers’ IDEs will automatically start using the custom model, which will inform all in-line code completions.

  5. Monitoring & Quality Assessment:
    Regularly retrain your custom model to keep it aligned with new code and evolving practices. Use the Copilot Usage Metrics API to track metrics like suggestion acceptance rates and see how much it’s improving.

Additional Resources

See more

Copilot Enterprise now helps you fix failed Actions jobs, plus other August updates (public beta)

Image

In this latest release, you can now ask Copilot Chat in GitHub.com questions about failed Actions jobs. With this feature, you can now speed up your pull request review cycle by asking Copilot about build failures to quickly get them resolved. In addition, we’ve added a quality improvement to how Copilot Chat in GitHub.com handles complex questions. This internal improvement will help you get the most out of your Copilot Chat conversations. Both of these features are in beta.

Copilot Chat in GitHub.com now has knowledge of failed Actions jobs

You can now click into a failed job on a pull request and ask Copilot what went wrong.

Open an existing PR and try it yourself:
– Tell me why this job failed
– Suggest a fix for this error

To learn more, check out our documentation.

Copilot Chat in GitHub.com can now answer complex questions

Copilot Chat can now access context from multiple primitives across pull requests, commits, discussions, issues, code, repos, and more to provide informed responses to more complex questions.

See it live by asking:
– How do I get started in this project?
– What are all of the open PRs assigned to me?
– Who can I talk to about this project?
– What changed on this PR?

We’re excited to bring these more advanced Copilot capabilities to customers in beta and would love your feedback!

How to enable these beta features for your enterprise

An enterprise owner can enable beta features using the Copilot policy “Opt in to preview features.”

Image

For more information about policies for Copilot Enterprise, see the documentation.

Join the discussion within the GitHub Community.

See more

Sign-up for the GitHub Copilot Extensions waitlist

Today, we are excited to open our waitlist for all GitHub Copilot users to start using Copilot Extensions!

Join the Copilot Extensions waitlist.

With extensions, you can extend the capabilities of GitHub Copilot Chat and enhance the experience to perform a wide range of actions across third-party tools, services, and data. Create feature flags, check log errors, access API documentation, and even deploy your application to the cloud, all through natural language.

Copilot Extensions are live on the GitHub Marketplace, with extensions from Octopus Deploy, Sentry, New Relic, and many more.

Questions or suggestions? Join the conversation in the community discussion.

See more

Copilot Metrics API Organization Team Metrics

We’re excited to share that usage metrics for GitHub Organization Teams are now available on the public beta of the GitHub Copilot Metrics API!

What metrics are available for GitHub Organization Teams?

  • Organization Team aggregates are available for teams with five or more Copilot license holders.
  • Teams must belong to the GitHub Organization which provisioned team members’ licenses.
  • The beta of the GitHub Copilot Metrics API is focused on serving metrics for Copilot Chat and code completions that take place in the IDE.
  • Code completion metrics include: Lines of Code Suggested, Lines of Code Accepted, Number of Suggestions, Number of Acceptances, and Active Users, with slices on language and IDE.
  • Copilot Chat metrics include: Number of Chats, Chat Suggestions Accepted, and Active Users. The endpoint does not currently feature slices on language or IDE for Chat metrics.

Documentation and Resources

See the following resources for help getting started:
– API Documentation: Explore the detailed API documentation, including metrics definitions here.
– Learning Pathway: You can find an extended article on measuring the impact of GitHub Copilot here.

Participate in the Public Beta!

Your feedback during this beta phase is invaluable to us. We encourage you to share your experiences, which will be instrumental in refining and enhancing the API as we look forward to the GA release.

Join the discussion within GitHub Community.

See more

Enterprise Team Metrics Now Available on the Copilot Metrics API

We’re happy to announce that metrics for GitHub Enterprise Teams are now available on the public beta of the GitHub Copilot Metrics API as of today.

The GitHub Copilot Metrics API is designed to supply you with information about Copilot’s usage within your organizations. The data from the API is intended to be consumed and combined with your organization’s own data to create greater visibility into how Copilot engagement fits into the bigger picture of your software development cycle.

What metrics are available for GitHub Enterprise Teams?

  • This iteration of the GitHub Copilot Metrics API is focused on serving metrics for Copilot Chat and code completions that take place in the IDE.
  • Code completion metrics include: Lines of Code Suggested, Lines of Code Accepted, Number of Suggestions, Number of Acceptances, and Active Users with slices on language, and IDE.
  • Copilot Chat metrics include: Number of Chats, Chat Suggestions Accepted, and Active Users. The endpoint does not currently feature slices on language or IDE for Chat metrics.
  • Enterprise Team-level aggregates are available for teams with five or more Copilot license holders.

Documentation and Resources

See the following resources for help getting started:
– API Documentation: Explore the detailed API documentation, including metrics definitions here.
– Learning Pathway: You can find an extended article on measuring the impact of GitHub Copilot here.

Participate in the Public Beta!

Your feedback during this beta phase is invaluable to us. We encourage you to share your experiences, which will be instrumental in refining and enhancing the API as we look toward the future.

Stay tuned for updates and enhancements throughout the beta period. We’re committed to delivering a robust and feature-rich API that meets your needs and expectations.

Join the discussion within GitHub Community.

See more

GitHub Copilot Chat and pull request summaries are now powered by GPT-4o

Copilot Chat and pull request summary generation now use GPT-4o, bringing the performance of OpenAI’s latest flagship model to all developers.

Copilot Chat is available in Visual Studio, VS Code, JetBrains IDEs, GitHub Mobile apps, and GitHub.com.

To use the new GPT-4o model in your IDE, ensure you are using at least the minimum version of Copilot Chat specified here:

What this means for Copilot users

With this upgrade to GPT-4o, Copilot users will experience the following benefits:

  1. Faster response times – up to 55% faster TTFT (time to first byte)
  2. More accurate and reliable Copilot Chat responses – our testing showed a 60% increase in user satisfaction.

Commitment to quality

The upgrade process focused on our unwavering commitment to quality, safety, and security. Here’s what that entailed:

  1. Offline and online evaluation: We performed rigorous offline and online testing to ensure the model brings tangible benefits to users. This involved thorough benchmarking and running simulations of real-world software development scenarios to validate the improved performance and accuracy of GPT-4o.
  2. Red teaming: To preemptively address any potential safety issues, we conducted extensive red teaming exercises. These tests challenged the model to ensure it meets our high standards for safety and reliability in diverse coding environments.

We can’t wait to see what you create with the new GPT-4o-powered Copilot!

Let us know your feedback and join the discussion within the GitHub Community.

Happy coding!

See more

Copilot Enterprise Mixed Licensing beta

Today, we’re introducing the beta for Copilot Enterprise Mixed Licensing within an enterprise. This grants GitHub Enterprise Cloud customers greater flexibility in selecting the best Copilot plans for their needs. Now, you can set a Copilot plan at the organization level instead of at the enterprise level.

Try it out now

To update an organization’s Copilot plan, an Enterprise Admin should navigate to Copilot Settings for the enterprise and select the desired plan via the dropdown menu for each organization.

Enterprise Mixed Licensing Dropdown Menu

Learn more about Copilot Enterprise Mixed Licensing in our documentation here and let us know what you think via Discussions.

See more

Copilot text completion for pull request descriptions beta

Enhance your pull request workflow: Copilot pull request text completion now in beta

Copilot text completion for pull request descriptions is now available to all Copilot Enterprise customers. After typing just a few characters, Copilot will suggest completions to finish your sentences, leveraging the context of the PR and linked issues to ensure highly accurate and relevant suggestions.

This feature is currently in beta. An enterprise or organization admin must enable beta features using the “Opt in to preview features” Copilot policy to access text completion.

animation of getting Copilot help writing a pull request description

How to enable text completion for your Enterprise

An enterprise admin can enable beta features using the Copilot policy.

screen grab of enterprise policy for enabling preview Copilot Enterprise features

For more information about policies for Copilot Enterprise, see the documentation.

Users can control the feature

This feature is on by default if you have a Copilot Enterprise seat and your organization has the “Opt in to preview features” policy enabled. Additionally, individual users have the ability to easily disable and reenable completions based on their personal preferences.

screen grab of user controlling text completion using the copilot menu in the pull request description

Learn more

To learn more, check the documentation for Copilot pull request text completion. This beta feature is subject to GitHub’s preview terms.

As always, we welcome any feedback on Copilot Enterprise in the discussion within GitHub Community.

See more

GitHub Copilot knowledge bases available in Visual Studio (preview)

GitHub Copilot Enterprise subscribers in Visual Studio can now use Copilot Chat to get answers enriched with context from Copilot knowledge bases. To try out this functionality, you’ll need to be running Visual Studio 17.11 Preview 3 or later.

You can access a knowledge base from any Copilot Chat conversation by typing @github, pressing the # key, selecting a knowledge base from the autocomplete, and then entering your question. Copilot will respond, using the Markdown documentation in your knowledge base as context for its answer.

For more details, check out the docs for Copilot Chat in Visual Studio. To try Copilot Chat yourself, download the extension.

See more

GitHub Copilot Enterprise on GPT-4o

GitHub Copilot Enterprise features on GitHub.com and Copilot Chat in GitHub Mobile are now powered by the latest model from OpenAI, GPT-4o.

See more

Copilot Enterprise knows about pull requests, discussions, and files – June Updates

June CE Changelog

Another month, and another exciting set of updates for Copilot Enterprise. Let’s dig in:

Copilot Chat in GitHub.com can now answer questions about your pull requests, discussions, and files.

Catch up on Pull Requests: Copilot can answer questions about a Pull Request and give you an overview of the changes the Pull Request introduces. To learn more, see “Asking questions about a specific pull request” in the GitHub docs.
– Try it yourself: Navigate to a Pull Request on GitHub.com, and ask Copilot to Tell me about this Pull Request

Get more out of Discussions Copilot can help you get up to speed quickly by summarizing discussions and discussion comments. In addition, Copilot can identify themes in the discussion and commentary made by different participants aiding you in catching up on context seamlessly. To learn more, see “Asking a question about a specific issue or discussion” in the GitHub docs.
– Try it yourself: Navigate to a discussion on GitHub.com, and ask Copilot to Summarize this discussion

Ask about Files and learn about recent changes: Copilot can now tell you about a file or retrieve the most recent changes in the file on any branch. Ask Copilot to tell you what has changed in a file to gain a deeper understanding into your codebase and what’s changing in it. To learn more, see “File details skill” in the GitHub docs.
– Try it yourself: Navigate to a file on GitHub.com, and ask Copilot to What's changed in this file recently?

Copilot Enterprise features are now available in Visual Studio Code and Visual Studio

As announced earlier this month, Copilot Enterprise features are now available in Visual Studio Code and in Visual Studio for the first time. (For Visual Studio, you’ll need to be running Visual Studio 17.11 Preview 2 or later.)

Chat with the context of the web in Visual Studio Code and Visual Studio: GitHub Copilot can now search Bing to find information outside of its general knowledge or your codebase. Just mention @GitHub, and Copilot will intelligently decide when to use Bing. Bing search is only available if enabled by an administrator – for more details, see “Enabling GitHub Copilot Enterprise features”.

Get answers from across your entire codebase in Visual Studio: Copilot Chat can now answer questions with understanding of your full repository, not just the tabs you have open. Index your repository on GitHub.com, and then ask a question mentioning @GitHub. You can ask questions like @GitHub Where is device detection implemented?.

Access your Copilot knowledge bases in Visual Studio Code: You can now access your knowledge bases from any Copilot Chat conversation by typing @GitHub #kb, selecting a knowledge base from the list, and then entering your question. Copilot will respond, using the Markdown documentation in your knowledge base as context for its answer.

As always, we welcome any feedback on Copilot Enterprise in the discussion within GitHub Community.

See more

New Copilot Enterprise features in VS Code (preview)

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.

To get started, make sure you’re running the latest release of the GitHub Copilot Chat extension for VS Code, and then check out the docs.

Access your Copilot knowledge bases

You can now access your knowledge bases from any Copilot Chat conversation in VS Code by typing @github #kb, selecting a knowledge base from the list, and then entering your question.

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

Search with the context of the web

GitHub Copilot can now search Bing within chat conversations in VS Code to answer questions and find information outside of its general knowledge or your codebase.

To get answers enriched with Bing search results, start your message with @github. Copilot will intelligently decide when to use Bing – or you can ask it to run a search by adding #web to your query. You can ask questions like @github What is the latest LTS version of Node.js?.

Bing search is only available if enabled by an administrator – for more details, see “Enabling GitHub Copilot Enterprise features”.

See more

New Copilot Enterprise features in Visual Studio (preview)

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.

To try out this functionality, install Visual Studio 17.11 Preview 2, and then check out the docs.

Get answers from across your entire codebase

Copilot Chat can now answer questions with understanding of your full repository, not just the tabs you have open. Index your repository on GitHub.com, and then ask a question mentioning @github. You can ask questions like @github Where is device detection implemented?.

Search with the context of the web

GitHub Copilot can now search Bing to find information outside of its general knowledge or your codebase. When you mention @github, Copilot will intelligently decide when to use Bing. You can ask questions like @github What are the breaking changes in Next.js v14?.

Bing search is only available if enabled by an administrator – for more details, see “Enabling GitHub Copilot Enterprise features”.

See more
View more changes