Since launching the public preview of GitHub Copilot coding agent for Jira, we’ve been listening closely to customer feedback. Thank you to everyone who has taken the time to try the integration, report issues, and share suggestions—your input has directly shaped the improvements we have shipped.

What’s new

Improved onboarding and setup guidance

Based on feedback from early adopters, the integration now provides better guidance, including clearer error messages and steps to resolve common configuration issues. We’ve also extended our documentation to more comprehensively cover integration prerequisites.

Model selection

You can now choose which AI model Copilot coding agent uses for your task, directly from Jira. Simply indicate the model you wish to use in your comment when mentioning @GitHub Copilot. For more information on available models, see Supported AI models in Copilot.

Jira ticket references in pull requests

We now instruct Copilot coding agent to include your Jira ticket number in the pull request title and branch name. The pull request will also include a link back to the originating Jira ticket and the Jira context provided to the agent. This makes it easier to trace work from Jira through to your codebase.

Confluence context via MCP

Provide Copilot coding agent with access to your Confluence pages by configuring the Atlassian MCP server with a personal access token (PAT). This allows the agent to reference design documents, specifications, and other context stored in Confluence when working on your Jira issues. For setup instructions, see Extending Copilot coding agent with MCP.

Getting started

To get all these updates, ensure you have the latest version of the GitHub Copilot for Jira app from your Atlassian instance.