Jira Extended MCP Server

Built by Moobean Team — a lightweight, Jira-only MCP server focused on getting things done with minimal setup.

What Makes This Different

• Jira-only, zero bloat — No Confluence, no extra modules. One uvx command and you're running.

• Wiki markup support — Uses Jira REST API v2, so *bold*, h2. Title, * bullet just work. No ADF JSON hassle.

• Bulk operations — Create up to 50 issues or transition multiple issues in a single call.

• Full release lifecycle — Create, update, delete versions and assign issues to releases.

• Issue links — Block, relate, duplicate, clone — with create, query, and delete support.

Related MCP server: Jira MCP Server

Comparison

Note: Feature data is based on each project's README and documentation as of March 2026. Features may have changed since then.

Use Cases

Just ask your AI agent in natural language:

Issue Management

"Create an epic in the KAN project titled 'User Auth System', start date April 1st, due date April 30th"

"Create 5 stories under KAN-42: Login, Sign Up, Password Reset, Social Login, 2FA"

"Show me all 'In Progress' issues in the KAN project"

Bulk Operations

"Transition all 10 backlog issues in this sprint to 'Done'"

"Show me the issue list included in the v2.0 release"

Releases & Sprints

"Create a v2.1.0 release in KAN project with release date May 15th"

"Move KAN-50 and KAN-51 to the current active sprint"

Issue Links

"Link KAN-10 as blocking KAN-20"

Rich Text (Wiki Markup)

"Create an issue with h2 headings and bullet lists in the description"

Quick Start

Prerequisites

• uv (Python package manager — installs Python automatically if needed)

• Jira API Token

Step 1: Install uv

uv is a Python package manager. If you don't have it:

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

# Windows (PowerShell)
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

Step 2: Get a Jira API Token

1. Go to https://id.atlassian.com/manage-profile/security/api-tokens

2. Click Create API token

3. Copy the token — you'll need it in the next step

Step 3: Configure your AI client

Choose where to add the config:

Easiest — one command:

# macOS / Linux
claude mcp add jira-extended -s user \
  -e JIRA_URL=https://your-instance.atlassian.net \
  -e JIRA_EMAIL=your-email@example.com \
  -e JIRA_API_TOKEN=your-token \
  -- uvx jira-extended-mcp

# Windows — use uvx.exe (not uvx) to avoid .cmd wrapper issues
claude mcp add jira-extended -s user \
  -e JIRA_URL=https://your-instance.atlassian.net \
  -e JIRA_EMAIL=your-email@example.com \
  -e JIRA_API_TOKEN=your-token \
  -- uvx.exe jira-extended-mcp

-s user installs globally. Omit it for project-only install.

Or edit the config file manually:

Open the file in a text editor:

# macOS / Linux
code ~/.claude.json    # or: nano ~/.claude.json

# Windows
notepad %USERPROFILE%\.claude.json

Add this content (create the file if it doesn't exist):

{
  "mcpServers": {
    "jira-extended": {
      "command": "uvx",
      "args": ["jira-extended-mcp"],
      "env": {
        "JIRA_URL": "https://your-instance.atlassian.net",
        "JIRA_EMAIL": "your-email@example.com",
        "JIRA_API_TOKEN": "your-api-token"
      }
    }
  }
}

Windows users: Use "command": "uvx.exe" instead of "command": "uvx". The uvx.cmd wrapper on Windows breaks the MCP stdio transport.

Open the config file in a text editor:

# macOS
code ~/Library/Application\ Support/Claude/claude_desktop_config.json

# Windows
notepad %APPDATA%\Claude\claude_desktop_config.json

Add or merge into the file:

{
  "mcpServers": {
    "jira-extended": {
      "command": "uvx",
      "args": ["jira-extended-mcp"],
      "env": {
        "JIRA_URL": "https://your-instance.atlassian.net",
        "JIRA_EMAIL": "your-email@example.com",
        "JIRA_API_TOKEN": "your-api-token"
      }
    }
  }
}

If the file already has other MCP servers, add the "jira-extended": {...} block inside the existing "mcpServers" object.

Windows users: Use "command": "uvx.exe" instead of "command": "uvx".

Create .vscode/mcp.json in your project root:

mkdir -p .vscode
code .vscode/mcp.json

{
  "servers": {
    "jira-extended": {
      "command": "uvx",
      "args": ["jira-extended-mcp"],
      "env": {
        "JIRA_URL": "https://your-instance.atlassian.net",
        "JIRA_EMAIL": "your-email@example.com",
        "JIRA_API_TOKEN": "your-api-token"
      }
    }
  }
}

Enable MCP: Settings > Chat > MCP must be checked. Works in Agent mode.

Windows users: Use "command": "uvx.exe" instead of "command": "uvx".

Open the config file:

# macOS / Linux
code ~/.cursor/mcp.json

# Windows
notepad %USERPROFILE%\.cursor\mcp.json

{
  "mcpServers": {
    "jira-extended": {
      "command": "uvx",
      "args": ["jira-extended-mcp"],
      "env": {
        "JIRA_URL": "https://your-instance.atlassian.net",
        "JIRA_EMAIL": "your-email@example.com",
        "JIRA_API_TOKEN": "your-api-token"
      }
    }
  }
}

Windows users: Use "command": "uvx.exe" instead of "command": "uvx".

git clone https://github.com/moobean-team/jira-extended-mcp-server.git
cd jira-extended-mcp-server
uv pip install -e .

Then use "command": "jira-extended-mcp" instead of "command": "uvx" in your config.

Step 4: Restart & verify

Restart your AI client, then ask:

"Show my Jira projects"

If you see your project list, you're all set.

Configuration

Environment Variables

Finding Your Start Date Field ID

The start date field ID varies per Jira instance. Use the get_createmeta tool for your project to see available fields, or:

curl -s -u email:token https://your-instance.atlassian.net/rest/api/2/field \
  | python -m json.tool | grep -i "start"

Rich Text (Wiki Markup)

This server uses Jira REST API v2, which accepts Jira wiki markup strings for description and comment fields. Jira renders them as rich text automatically.

Full reference: Jira Wiki Markup

Available Tools

Architecture

src/jira_extended_mcp/
├── server.py    # FastMCP server + 27 tool definitions
├── client.py    # Async Jira REST client (httpx + rate limit retry)
├── adf.py       # ADF fallback helpers (v3 response parsing)
└── __init__.py

Key design decisions:

Development

git clone https://github.com/moobean-team/jira-extended-mcp-server.git
cd jira-extended-mcp-server
uv pip install -e .

# Run directly
jira-extended-mcp

# Or via module
python -m jira_extended_mcp.server

Troubleshooting

Ensure JIRA_URL, JIRA_EMAIL, and JIRA_API_TOKEN are set in your MCP config's env block. The server checks these on startup.

Jira transitions are workflow-specific. Use get_transitions first to see available transitions for the issue's current status. Transition names are case-insensitive.

Your Jira instance may use a different custom field ID. Use get_createmeta to find the correct field, then set JIRA_START_DATE_FIELD env var.

The server automatically retries up to 3 times using the Retry-After header. For bulk operations with 50+ issues, consider splitting into multiple calls.

This server uses Jira REST API v2 which accepts wiki markup. Use Jira wiki syntax (*bold*, h2. Title, * bullet) instead of Markdown.

License

MIT © Moobean Team