Jira

You can manage Jira issues programmatically through a Model Context Protocol (MCP) server built to integrate with Jira’s REST API. This server lets you create, list, update, retrieve, delete, and comment on Jira issues from your AI-powered workflows, streamlining project management inside your automation pipelines.

How to use

To use this MCP server with your MCP client, connect to the local MCP command that runs the Jira integration. Once connected, you can perform common Jira operations as actions in your workflows: create issues (Task, Epic, Subtask), list issues with optional status filtering, update fields such as summary or description, fetch complete details for an issue, delete issues, and add comments. The server formats issue information consistently, including key details like key, summary, type, status, creation info, description, and comments.

In practice, you would configure your MCP client to invoke the Jira MCP server’s endpoints in your automation scripts. Each operation corresponds to a clear action you can call from your workflow, enabling seamless Jira management within your AI-assisted processes.

How to install

Prerequisites you need before installing include: a Jira account with API access and a Jira API token. Follow these concrete steps to install and run the Jira MCP server.

Install via Smithery (automatic installation):

npx -y @smithery/cli install jira-server --client claude

Manual installation steps (local development):

npm install

npm run build

Configuration

Create a Jira config to specify your projects and Jira connection details, which the MCP server uses to locate and manage issues.

{
  "projectKey": ["YOUR_PROJECT_KEY", "YOUR_OTHER_PROJECT_KEY"]
}

Configure the MCP server to run with your Jira credentials and domain. Place this configuration in your desktop or system config area as shown, replacing the placeholders with your actual values.

{
  "mcpServers": {
    "jira": {
      "command": "node",
      "args": ["/path/to/jira-server/build/index.js"],
      "env": {
        "JIRA_EMAIL": "your-email@example.com",
        "JIRA_API_TOKEN": "your-api-token",
        "JIRA_DOMAIN": "your-domain"
      }
    }
  }
}

Security and maintenance notes

Protect your Jira API token as a sensitive credential. Store it in a secure location and restrict access to systems and people who require it. Regularly rotate tokens and keep dependencies up to date to mitigate security risks.

If you need to update the server or its dependencies, use the standard build and restart flow to ensure changes take effect in your MCP environment.

Usage patterns and tips

Use the create, list, update, get, delete, and add_comment operations to automate Jira workflows. For example, you can automatically create tasks when a new issue is requested in your AI assistant, filter lists by status to monitor progress, or add daily summary comments to active issues to keep stakeholders informed.

Available tools

create_issue

Creates a new Jira issue with a title, description, and type (Task, Epic, or Subtask).

list_issues

Lists issues for a configured project, with optional status filtering.

update_issue

Updates an existing issue’s summary, description, or status.

get_issue

Retrieves detailed information about a specific issue.

delete_issue

Deletes a Jira issue by key.

add_comment

Adds a comment to an existing issue.