I have already connected Claude with Jira, and the setup was quite straightforward using the Atlassian connector in Claude, which uses MCP.
My expectation for the Cursor–Jira integration is the following: Cursor would have access to the current codebase, and through the Jira integration, it could fetch the details of a specific Jira ticket describing the required changes. Cursor would then use the ticket details along with the existing code to understand the requirement and suggest the necessary code changes.
The Jira access would ideally be read-only—Cursor should be able to fetch ticket details but not modify Jira tickets.
Please correct me if I have misunderstood how the Cursor–Jira integration works or if there is a better approach to achieve this.
Thank you
Mahanth