JIRA MCP Server (Async)

@judexzhu

About JIRA MCP Server (Async)

A high-performance, asynchronous Model Context Protocol (MCP) server that integrates with JIRA using stdio transport

#mcp-jira #jira-integration #async-server

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "mcp-jira-judexzhu": {
      "command": "uv",
      "args": [
        "sync"
      ]
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is JIRA MCP Server (Async)?

A high-performance, asynchronous Model Context Protocol (MCP) server that integrates with JIRA using stdio transport. It allows AI assistants to perform JIRA operations—search, create, update issues, and track relationships—via MCP tools. Designed for developers using MCP clients like Claude Desktop.

How to use JIRA MCP Server (Async)?

Install with uv sync, configure environment variables JIRA_SERVER_URL and JIRA_API_TOKEN, then add the server to your MCP client’s configuration (e.g., Claude Desktop’s mcpServers block). The server runs over stdio and exposes tools with jira_ prefixes.

Key features of JIRA MCP Server (Async)

8 MCP tools covering search, read, and write operations
Asynchronous, non-blocking API calls with connection pooling
Built-in rate limiting to respect JIRA API limits
Uses stdio transport for MCP client integration
Environment variables for timeouts, logging, and concurrency
Clean, tool-focused architecture without MCP resources

Use cases of JIRA MCP Server (Async)

Search JIRA issues with JQL from an AI assistant
Retrieve detailed issue information, comments, and links
Create new issues or update existing ones programmatically
Track epics, subtasks, and workflow transitions

FAQ from JIRA MCP Server (Async)

What JIRA operations does the server support?

The server provides tools for searching issues (jira_search_issues), getting details (jira_get_issue_details), reading comments and links, managing epics/subtasks, viewing transitions, and creating/updating issues—all with a jira_ tool prefix.

What environment variables are required?

JIRA_SERVER_URL (your JIRA instance URL) and JIRA_API_TOKEN are required. Optional variables include MAX_CONCURRENT_REQUESTS (default 2), REQUEST_TIMEOUT (30s), LOG_LEVEL, and more.

What transport does the server use?

It uses stdio transport, optimized for MCP client integration. No HTTP endpoints are exposed; the server communicates over standard input/output.

What are the runtime dependencies?

Python 3.13+ and the uv package manager. The server uses aiohttp for async HTTP calls and implements built-in connection pooling and rate limiting.

Are there any known limitations?

Rate limiting is built-in; the default MAX_CONCURRENT_REQUESTS value is 2 (requests per second). The server does not expose MCP resources—only tools—and relies entirely on the provided API token for authentication.

Frequently asked questions

What JIRA operations does the server support?

The server provides tools for searching issues (`jira_search_issues`), getting details (`jira_get_issue_details`), reading comments and links, managing epics/subtasks, viewing transitions, and creating/updating issues—all with a `jira_` tool prefix.

What environment variables are required?

`JIRA_SERVER_URL` (your JIRA instance URL) and `JIRA_API_TOKEN` are required. Optional variables include `MAX_CONCURRENT_REQUESTS` (default 2), `REQUEST_TIMEOUT` (30s), `LOG_LEVEL`, and more.

What transport does the server use?

It uses stdio transport, optimized for MCP client integration. No HTTP endpoints are exposed; the server communicates over standard input/output.

What are the runtime dependencies?

Python 3.13+ and the `uv` package manager. The server uses `aiohttp` for async HTTP calls and implements built-in connection pooling and rate limiting.

Are there any known limitations?

Rate limiting is built-in; the default `MAX_CONCURRENT_REQUESTS` value is 2 (requests per second). The server does not expose MCP resources—only tools—and relies entirely on the provided API token for authentication.