Atlassian Dc Mcp Jira
Connect Claude Code or Cursor to self-hosted Jira Data Center so the agent can search, read, and create issues without leaving the editor.
Overview
Atlassian DC MCP Jira is an MCP server for the Build phase that lets AI coding agents search, view, and create issues in Atlassian Jira Data Center through stdio and REST API credentials.
What is this MCP server?
- Search Jira Data Center issues from natural-language agent requests
- View full issue details without context-switching to the Jira UI
- Create new issues from the agent during planning or bug triage
- Stdio MCP transport via npm package @atlassian-dc-mcp/jira (v0.9.9)
- Configure with JIRA_HOST or JIRA_API_BASE_PATH plus required JIRA_API_TOKEN
- Package version 0.9.9 on npm as @atlassian-dc-mcp/jira
- 3 documented environment variables: JIRA_HOST, JIRA_API_BASE_PATH, JIRA_API_TOKEN (token required)
- Repository: github.com/b1ff/atlassian-dc-mcp
What problem does it solve?
Switching between your IDE and Jira to look up tickets or log work breaks agent flow and wastes time on self-hosted Jira Data Center setups.
Who is it for?
Solo builders or tiny teams on Jira Data Center who want MCP-driven issue lookup and creation from Claude Code, Cursor, or Codex during development.
Skip if: Builders without Jira Data Center access, those who need full board/sprint admin only in-browser, or workflows that require Confluence or Bitbucket from the same repo without those MCP packages.
What do I get? / Deliverables
After you register the server with host and API token, your agent can query and create Jira issues in conversation without opening the Jira UI.
- Agent-callable Jira issue search and view against Data Center
- Agent-driven creation of new Jira issues
- Stdio MCP server registration using published server schema v0.9.9
Recommended MCP Servers
Journey fit
Builders install this while actively developing and tracking work in Jira; the canonical shelf is Build because the value is agent-assisted issue ops tied to implementation, not launch SEO or ops monitoring alone. It is an MCP stdio bridge to Jira’s REST API with token-based auth, which matches the integrations subphase rather than generic docs or frontend work.
How it compares
Jira Data Center MCP integration, not a standalone planning skill or a curated skills marketplace.
Common Questions / FAQ
Who is Atlassian DC MCP Jira for?
It is for developers and solo builders who use self-hosted Jira Data Center and want their AI coding agent to search, view, and create issues via MCP.
When should I use Atlassian DC MCP Jira?
Use it during build and PM workflows when you need ticket context, backlog search, or quick issue creation from the agent without leaving your editor.
How do I add Atlassian DC MCP Jira to my agent?
Install @atlassian-dc-mcp/jira from npm, set JIRA_API_TOKEN (required) and JIRA_HOST or JIRA_API_BASE_PATH, then register the stdio MCP server in Claude Code, Cursor, or your client’s MCP config.