jira-mcp

MCP server for fetching Jira issue data.

Required Settings

Python

This project requires Python 3.14 or newer.

The required version is declared in:

• .python-version
• pyproject.toml
• uv.lock

Dependency Management

Use uv as the Python dependency and virtual environment manager for this project.

Install uv if it is not already available:

curl -LsSf https://astral.sh/uv/install.sh | sh

Then install the project dependencies from pyproject.toml and uv.lock:

uv sync

Do not install dependencies directly with pip; keep dependency changes in pyproject.toml and refresh the lockfile with uv.

Environment Variables

Create a local .env file from the example file:

cp .env.example .env

Set these values in .env:

JIRA_EMAIL=your-email@example.com
JIRA_API_TOKEN=your-jira-token
JIRA_DOMAIN=your-company.atlassian.net

Required by the current code:

• JIRA_EMAIL: Email address used for Jira API authentication.
• JIRA_API_TOKEN: Jira API token for the account above.
• JIRA_DOMAIN: Jira site domain, for example your-company.atlassian.net.

Included in .env.example but not currently used by main.py:

• MODEL_NAME
• OPENAI_API_KEY

Jira Access

The Jira API call uses:

https://<JIRA_DOMAIN>/rest/api/3/issue/<ISSUE_KEY>

The Jira account must have permission to view the requested issue. For Atlassian Cloud, use an API token rather than your account password.

Running

Run the MCP server over stdio:

uv run main.py

The server exposes the fetch_jira MCP tool.

Cursor MCP Setup

To use this MCP server from another project in Cursor, create or update this file in that project:

.cursor/mcp.json

Add the Jira MCP server configuration:

{
  "mcpServers": {
    "jira-server": {
      "command": "uv",
      "args": [
        "--directory",
        "/path/to/jira-mcp",
        "run",
        "main.py"
      ]
    }
  }
}

Replace /path/to/jira-mcp with the local path to this project on your computer.