JIRA MCP Server

A simple Model Context Protocol (MCP) "vibe-coded" server for integrating JIRA with Cursor IDE. MCP is an open protocol that enables seamless integration between LLM applications and external data sources and tools.

This implementation started out by discarding another JIRA MCP server that we failed to init in Cursor.

BEWARE! Even this document is almost entirely written by AI coding assistant.

Features

Get JIRA issues by key
Search issues using JQL (JIRA Query Language)
Create and update issues
Clone issues (helpful to work around mandatory custom fields)
Configurable field selection
Pagination support
Detailed error handling and logging
Log work

User Workflows

Search and Filter Flow

Issue Cloning Flow

For detailed technical architecture and system workflows, including issue lifecycle and authentication flows, see ARCHITECTURE.md.

About MCP

This server implements the Model Context Protocol specification, allowing Cursor IDE to seamlessly interact with JIRA data through its AI features. The protocol standardizes how LLM applications communicate with external data sources and tools.

Setup

Create a virtual environment:

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

Install dependencies:

pip install -r requirements.txt

Configure environment variables: Create a .env file with:

JIRA_URL=your_jira_url
JIRA_USERNAME=your_username
JIRA_API_TOKEN=your_api_token

Usage

Run the server:

./run-jira-mcp.sh

Development

The project follows the implementation plan outlined in IMPLEMENTATION_PLAN.md.

Current version: v0.3

✅ Basic JIRA integration
✅ Search functionality with JQL support

Model Context Protocol - The main MCP project
MCP Python SDK - The SDK we use to implement this server
MCP Documentation - Protocol documentation and specifications

License

MIT

Integrations