How to Set Up Jira MCP Server for AI Coding (2026)

How to Set Up Jira MCP Server for AI Coding in 2026

If you use Jira for project management and AI coding assistants like Claude Code or Cursor for development, connecting them through a Model Context Protocol (MCP) server can dramatically streamline your workflow. Instead of copy-pasting ticket descriptions into your AI assistant, the MCP server lets your AI read Jira issues, pull acceptance criteria, and even update ticket statuses directly.

This guide walks you through setting up a Jira MCP server from scratch, configuring it with popular AI coding tools, and getting the most out of the integration.

What Is MCP and Why Does It Matter?

The Model Context Protocol (MCP) is an open standard that allows AI models to interact with external tools and data sources. Think of it as a universal adapter between your AI assistant and the services you already use.

For developers, MCP means your AI coding assistant can:

• Read Jira tickets to understand requirements before writing code
• Search issues by project, sprint, or custom JQL queries
• Update ticket statuses when work is completed
• Add comments to tickets with implementation notes
• Pull acceptance criteria directly into the coding context

Prerequisites

Before you begin, make sure you have:

1. A Jira Cloud or Jira Data Center instance with API access
2. A Jira API token (generate one at id.atlassian.com)
3. Node.js 18+ installed on your machine
4. An MCP-compatible AI tool such as Claude Code, Cursor, or Windsurf

Step 1: Generate a Jira API Token

Navigate to your Atlassian account settings and create a new API token:

1. Go to https://id.atlassian.com/manage-profile/security/api-tokens
2. Click Create API token
3. Give it a descriptive label like mcp-server-jira
4. Copy the token immediately (you will not see it again)

Store the token securely. You will need it alongside your Jira email and instance URL.

Step 2: Install the Jira MCP Server

The most popular open-source Jira MCP server is @anthropic/mcp-server-jira. Install it globally:

npm install -g @anthropic/mcp-server-jira

Alternatively, you can use npx to run it without a global install:

npx @anthropic/mcp-server-jira

If you prefer a Docker-based setup:

docker run -d \
  -e JIRA_URL=https://yourcompany.atlassian.net \
  -e JIRA_EMAIL=your-email@company.com \
  -e JIRA_API_TOKEN=your-api-token \
  -p 3100:3100 \
  ghcr.io/anthropic/mcp-server-jira:latest

Step 3: Configure Environment Variables

Create a .env file in your project root or export the variables in your shell:

export JIRA_URL="https://yourcompany.atlassian.net"
export JIRA_EMAIL="your-email@company.com"
export JIRA_API_TOKEN="your-api-token-here"
export JIRA_PROJECT_KEY="PROJ"  # Optional: default project

Step 4: Connect to Claude Code

To use the Jira MCP server with Claude Code, add the following to your ~/.claude/mcp_servers.json configuration:

{
  "jira": {
    "command": "npx",
    "args": ["@anthropic/mcp-server-jira"],
    "env": {
      "JIRA_URL": "https://yourcompany.atlassian.net",
      "JIRA_EMAIL": "your-email@company.com",
      "JIRA_API_TOKEN": "your-api-token-here"
    }
  }
}

Restart Claude Code, and the Jira tools will appear in your tool list. You can verify by typing /mcp in the Claude Code CLI.

Step 5: Connect to Cursor

For Cursor, add the MCP server in your .cursor/mcp.json file:

{
  "mcpServers": {
    "jira": {
      "command": "npx",
      "args": ["@anthropic/mcp-server-jira"],
      "env": {
        "JIRA_URL": "https://yourcompany.atlassian.net",
        "JIRA_EMAIL": "your-email@company.com",
        "JIRA_API_TOKEN": "your-api-token-here"
      }
    }
  }
}

Step 6: Test the Connection

Once configured, test the integration by asking your AI assistant to fetch a Jira issue:

> Fetch the details of PROJ-1234 from Jira

The AI should return the ticket summary, description, status, assignee, and any subtasks.

Common MCP Tools Exposed by Jira Server

Tool Name	Description	Example Usage
jira_get_issue	Fetch a single issue by key	"Get details of PROJ-123"
jira_search	Search issues using JQL	"Find all open bugs in Sprint 5"
jira_create_issue	Create a new Jira issue	"Create a bug ticket for login failure"
jira_update_issue	Update issue fields	"Move PROJ-123 to In Review"
jira_add_comment	Add a comment to an issue	"Comment on PROJ-123 with implementation notes"
jira_list_projects	List available projects	"Show me all Jira projects"
jira_get_sprint	Get active sprint info	"What tickets are in the current sprint?"

Practical Workflow: AI-Driven Development with Jira

Here is a real-world workflow that combines Jira MCP with AI coding:

1. Ask the AI to pull the next ticket: "What is the highest priority unassigned ticket in our current sprint?"
2. Let the AI read requirements: The assistant fetches the full ticket description, acceptance criteria, and linked issues.
3. Generate code based on the ticket: "Implement the feature described in PROJ-456."
4. Update the ticket: "Move PROJ-456 to In Review and add a comment summarizing the changes."

This loop eliminates constant context-switching between Jira and your IDE.

Troubleshooting Common Issues

Authentication errors (401) Double-check your API token and email. Jira Cloud requires the email associated with the token, not your username.

Permission denied on specific projects Ensure the Jira account associated with your API token has access to the projects you are querying.

MCP server not appearing in tool list Restart your AI coding tool after modifying the MCP configuration. Check the server logs for startup errors:

npx @anthropic/mcp-server-jira 2>&1 | tee mcp-jira.log

Slow responses on large Jira instances Add JQL filters to narrow search scope. Instead of searching all issues, restrict to specific projects or sprints.

Comparison: Jira MCP Servers

Feature	@anthropic/mcp-server-jira	mcp-atlassian	Custom Build
Setup difficulty	Easy	Medium	Hard
Jira Cloud support	Yes	Yes	Depends
Jira Data Center	Limited	Yes	Depends
JQL search	Yes	Yes	Depends
Issue creation	Yes	Yes	Depends
Confluence integration	No	Yes	Depends
Maintenance	Community	Community	You

Security Best Practices

• Never commit API tokens to version control. Use environment variables or a secrets manager.
• Use a dedicated Jira service account with minimal permissions rather than your personal account.
• Rotate API tokens regularly, especially if shared across team members.
• Restrict project access on the Jira side to limit what the MCP server can see.

Conclusion

Setting up a Jira MCP server bridges the gap between your project management workflow and AI-powered development. Once connected, your AI coding assistant becomes context-aware, understanding your tickets, requirements, and sprint goals without manual copy-pasting.

If you are building AI-powered applications and need reliable media generation APIs for images, video, and audio, check out Hypereal AI. Hypereal provides a unified API for AI media generation with competitive pricing, no content restrictions on creative work, and fast inference times -- perfect for developers integrating AI capabilities into their products.