Jira MCP Server: AI-Powered Project Management

Every engineering team lives in Jira, but nobody enjoys navigating it. The Jira MCP server brings your entire backlog into Claude — search for issues, create tickets, update status, and query sprint progress all through a conversation. Instead of clicking through boards and filters, you tell Claude what you need.

What is Jira?

Jira is Atlassian's project management and issue tracking platform, used by software teams worldwide to plan, track, and ship work. It organizes work into issues — bugs, stories, tasks, and epics — grouped into sprints for agile teams or managed in kanban boards for continuous flow. A typical team uses Jira to maintain a backlog, run two-week sprints, track bug reports, manage releases, and report on velocity and delivery. Despite its ubiquity, Jira's interface can be slow and complex for routine lookups, which is exactly where the MCP server adds value.

What the Jira MCP Server Does

The Jira MCP server connects Claude to the Atlassian REST API, exposing your project data as tools Claude can call during a conversation. Once installed, Claude can:

• Search issues with JQL — run any Jira Query Language query and get structured results back in your conversation.
• Create new issues — specify summary, description, issue type, priority, and assignee in plain English; Claude handles the API call.
• Update status, assignee, and priority — transition issues through your workflow states and reassign or reprioritize without opening Jira.
• List sprint items — get all issues in the current active sprint for any project, filtered by status or assignee if needed.
• Get issue details and comments — retrieve the full context of any ticket, including description, comments, attachments metadata, and linked issues.
• Add comments to tickets — post comments to any issue directly from your Claude conversation, with full Jira markdown support.

Claude constructs the correct JQL queries and API payloads based on your plain-English requests, so you never need to remember Jira's query syntax. To understand how Claude decides which Jira tools to use, see our claude_desktop_config.json guide for a full picture of how MCP servers are registered and invoked.

Prerequisites

Before installing the Jira MCP server, make sure you have the following:

• Jira account — Jira Cloud is recommended for the best API compatibility. Jira Data Center works but may require additional configuration.
• Atlassian API token — go to id.atlassian.com/manage-profile/security/api-tokens and create a new token. This token authenticates Claude to the Jira API on your behalf.
• Jira base URL — this is the domain of your Jira instance, typically in the format https://yourcompany.atlassian.net. Find it in your browser's address bar when you are logged into Jira.
• Node.js 18 or higher — check with node --version. Download from nodejs.org if needed.
• Claude Desktop — the MCP server connects via Claude Desktop's MCP integration.

How to Install the Jira MCP Server

1. Get your Atlassian API token. Visit id.atlassian.com/manage-profile/security/api-tokens, click "Create API token", give it a name like "Claude MCP", and copy the token. Store it somewhere safe — you will not be able to view it again after closing the dialog.
2. Find your Jira base URL. Log into your Jira instance and look at the address bar. The base URL is everything up to and including .atlassian.net, for example https://yourcompany.atlassian.net.
3. Open your Claude Desktop config file. This file is claude_desktop_config.json. On macOS it lives at ~/Library/Application Support/Claude/claude_desktop_config.json; on Windows it is at %APPDATA%\Claude\claude_desktop_config.json. See our full config guide for details.
4. Add the Jira server block with your base URL, email address, and API token, then save the file and restart Claude Desktop. The Jira tools will appear in Claude's tool list within a few seconds.

The Config JSON

Add this block to your claude_desktop_config.json. Replace all three placeholder values with your actual Jira base URL, account email, and API token.

{
  "mcpServers": {
    "jira": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-jira"],
      "env": {
        "JIRA_BASE_URL": "https://yourcompany.atlassian.net",
        "JIRA_USERNAME": "your@email.com",
        "JIRA_API_TOKEN": "your_api_token_here"
      }
    }
  }
}

Your API token gives Claude access to everything your Jira account can see. For guidance on scoping access and keeping credentials secure, read our MCP security and trust levels guide before going further.

Real Use Cases

Once the server is connected, these are the kinds of requests Claude can handle immediately:

1. "Show me all open bugs in the current sprint." Claude constructs a JQL query for your active sprint filtered to bug-type issues with open status, and returns a table with ticket keys, summaries, priorities, and assignees.
2. "Create a ticket for the login page bug with high priority." Claude calls the Jira create issue API with the summary you describe, sets the issue type to Bug, sets priority to High, and returns the new ticket key so you can share it immediately.
3. "What did the backend team close last week?" Claude queries for issues resolved in the past 7 days filtered to backend team members, giving you a summary of completed work without logging into Jira or building a custom filter.
4. "Move PROJ-123 to In Progress and assign it to me." Claude calls the transition API to move the issue to the correct workflow state and updates the assignee to your account — two API calls handled in one natural-language request.
5. "List all tickets blocking our release." Claude searches for issues tagged with a blocker link type or a specific fix version, returning every ticket that needs resolution before your release can ship.

Pros

• JQL support for powerful queries — Claude can construct complex Jira Query Language expressions from simple English, unlocking Jira's full search power without you needing to learn JQL syntax.
• Create and update tickets without leaving Claude — the full ticket lifecycle — creation, assignment, transitions, and comments — is available in your conversation, eliminating context switching.
• Combine with GitHub MCP for a full dev workflow — pair the Jira MCP server with the GitHub MCP server to link tickets to pull requests, check commit status against issues, and manage your entire development cycle from a single Claude conversation.

Cons / Limitations

• Jira API rate limits — the Atlassian REST API enforces rate limits based on your Jira plan. High-frequency queries or bulk operations may hit these limits; space out requests if you are doing large-scale operations.
• Advanced features need configuration — complex Jira features like custom workflows, custom fields, and automation rules may require additional configuration beyond the basic server setup. The server works best with standard Jira configurations.
• Read/write permissions need careful scoping — your API token inherits the full permissions of your Jira account. If Claude can update tickets, it can update any ticket your account has access to. Consider creating a dedicated service account with scoped permissions for production use.