Most Jira servers give you issues and leave the sprint maths to you. Here burndown, velocity, goal progress and blockers are each their own call, and the dashboard tool bundles them — so 'how is the sprint going' is one question rather than four JQL queries and a spreadsheet.
A Jira server built around sprint analytics rather than issue CRUD. It reads burndown, velocity, goal progress and blocked issues, and can assemble them into one dashboard. Authentication runs through a browser OAuth flow so nobody has to go looking for an API token.
- get_sprint_burndown — burndown for the current sprint
- get_team_velocity — velocity across recent sprints
- get_sprint_goal_progress — how the sprint goal itself is tracking, not just the ticket count
- get_blocked_issues — what is stuck, gathered rather than hunted for
- generate_dashboard — all of the above assembled in one call
- jira_search with JQL, and jira_get_issue for a single issue's detail
- list_projects for what the account can actually reach
- start_oauth and oauth_status to run and check the browser login
- test_jira_connection to confirm the connection before you trust a number
A Jira Cloud site and an Atlassian account that can see the boards you care about. The hosted path takes your Jira URL and email in the marketplace config and then runs the browser login; scopes cover reading users and work and writing work. Self-hosting means registering your own OAuth app with Atlassian and supplying OAUTH_CLIENT_ID and OAUTH_CLIENT_SECRET. Node.js 18 or newer. Tokens are held in memory, so a restart means signing in again.