概要
What is Jira MCP Server?
A Model Context Protocol (MCP) server that enables AI assistants like Claude to interact with Jira, allowing users to manage projects, issues, and sprints through natural language commands.
How to use Jira MCP Server?
Clone the repository, install dependencies with npm install, build with npm run build, then configure the MCP server in your cursor/windsurf settings using environment variables for Jira host, email, and API token. Run the server via node ./dist/index.js.
Key features of Jira MCP Server
- List all accessible Jira projects
- Get detailed issue information by key
- Search issues by project and assignee
- List project members and check user membership
- Create new issues with custom fields
- List and query sprints with filtering
Use cases of Jira MCP Server
- Ask an AI assistant to show all Jira projects
- Retrieve and inspect details of a specific issue
- Find issues assigned to a team member in a project
- Create a new bug or task without leaving the chat
- Check active sprints for a board or project
FAQ from Jira MCP Server
What does Jira MCP Server do?
It acts as a bridge between MCP-compatible AI assistants (like Claude) and Jira, providing tools to list projects, get issues, search, create issues, and manage sprints.
What dependencies are required?
Node.js (TypeScript), and the npm package dependencies installed via npm install. The server uses MCP protocol.
How do I authenticate with Jira?
You need a Jira API token, your Atlassian account email, and your Jira domain host (e.g., your-domain.atlassian.net). Add these as environment variables in the MCP server configuration.
Where does my data stay?
All data stays within your Jira instance (Atlassian cloud). The server makes API calls to Jira on your behalf; no data is stored locally beyond temporary processing.
What transport does the server use?
The server communicates via the MCP protocol using standard input/output (stdin/stdout) when launched with the node command.
