Overview
What is Jira MCP Server?
Jira MCP Server is a Model Context Protocol server that integrates with Jira, enabling LLMs to interact with Jira projects, boards, sprints, and issues via its REST API.
How to use Jira MCP Server?
Install via npx with environment variables JIRA_PERSONAL_ACCESS_TOKEN and JIRA_BASE_URL. Configure in Claude Desktop’s claude_desktop_config.json. Requires Node version 22.12.0 or above.
Key features of Jira MCP Server
- Lists Jira projects with optional filters and expand parameters.
- Lists boards from a project by key or ID.
- Lists sprints from a specific board.
- Lists issues from a given sprint.
Use cases of Jira MCP Server
- Ask an LLM “Show me all Jira projects” to retrieve the project list.
- Inquire about Kanban boards in a specific project.
- Request all sprints for a given board ID.
- Retrieve issues in a particular sprint with optional pagination.
FAQ from Jira MCP Server
What are the runtime requirements?
Node version 22.12.0 or above. A Jira Personal Access Token and Jira base URL must be provided as environment variables.
How is authentication handled?
Authentication uses a Jira Personal Access Token in the format [email protected]:your_personal_jira_access_token passed via the JIRA_PERSONAL_ACCESS_TOKEN environment variable.
Where does the data live?
All data resides in your Jira instance. The server queries your Jira via the REST API and does not store any data locally.
What are the known limits?
The maxResults parameter for tools is capped at 100. Pagination is supported via startAt.
What transport does the server use?
The server uses standard MCP transport (stdio) as configured in Claude Desktop’s configuration.
