Jira MCP Server

What is Jira MCP Server?

A Model Context Protocol (MCP) server that allows AI assistants to interact with Jira. It enables AI to fetch user's assigned tickets, add comments, get available status transitions, and update ticket status. Designed for developers who want to give AI agents Jira integration capabilities.

How to use Jira MCP Server?

Set up a .env file with JIRA_SERVER, JIRA_USERNAME, JIRA_API_KEY, and JIRA_PROJECT_KEY. Then run either the Docker script (./run_docker_locally.sh) or install Python dependencies and start with uvicorn src.server:app --reload. The server runs on http://localhost:8000 and the MCP endpoint is at /mcp.

Key features of Jira MCP Server

• Fetch user's assigned tickets
• Add comments to Jira tickets
• Get available status transitions
• Update ticket status
• Simple environment variable configuration
• Runs locally via Docker or Python

Use cases of Jira MCP Server

• AI assistant fetching your open tickets
• Automating ticket comments from chat
• Moving tickets through workflow statuses
• Integrating AI workflows with Jira project management
• Allowing conversational Jira updates

FAQ from Jira MCP Server

What environment variables are required?

JIRA_SERVER (your Atlassian instance URL), JIRA_USERNAME (your email), JIRA_API_KEY, and JIRA_PROJECT_KEY. Create a .env file with these values.

How do I get my Jira API key?

Go to https://id.atlassian.com/manage-profile/security/api-tokens and generate a token. Use that as JIRA_API_KEY in your .env file.

What are the runtime requirements?

The server requires Docker (for the Docker option) or Python with the dependencies listed in requirements.txt. You also need a Jira account with API access.

Can this server modify ticket statuses?

Yes, it can get available status transitions and update ticket status accordingly.

Where is the server exposed?

It starts on http://localhost:8000. The MCP endpoint is at http://localhost:8000/mcp.