@abdelrahmanhsn/jira-mcp
A Model Context Protocol server that connects your AI IDE to Jira. Query your tickets, active sprint, and issue details directly from GitHub Copilot, Cursor, Claude Desktop, or any MCP-compatible client.
README
@abdelrahmanhsn/jira-mcp
A Model Context Protocol (MCP) server that connects your AI IDE to Jira. Query tickets, manage sprints, and let AI autonomously implement, test, and ship Jira tickets — all without leaving your editor.
Works with GitHub Copilot, Cursor, Claude Desktop, and any MCP-compatible client.
Tools
| Tool | Description |
|---|---|
get_my_tickets |
Get all Jira tickets assigned to you, ordered by last updated |
get_active_sprint_tickets |
Get your tickets in the currently active sprint |
get_issue_details |
Get full details (description + attachments) for a specific issue key |
add_comment |
Add a comment to any Jira issue |
get_my_standup |
Get a standup summary of tickets you updated since yesterday |
get_sprint_summary |
Get all sprint tickets grouped by status (Todo / In Progress / Done) |
search_tickets |
Search tickets with plain English or raw JQL |
get_context_for_pr |
Extract Jira ticket from a branch name and return a ready-to-use PR description block |
start_ticket |
🤖 Autonomous mode — assigns ticket, moves to In Progress, creates git branch, then drives AI to implement, test, commit, push, open PR, and comment on Jira — non-stop |
Prerequisites
- Node.js 18 or later
- A Jira Cloud account
- A Jira API token (generate one here)
- GitHub CLI — required for
start_ticketto create PRs automatically
Install GitHub CLI
# macOS
brew install gh
gh auth login
# Windows
winget install --id GitHub.cli
gh auth login
# Linux
# See https://github.com/cli/cli/blob/trunk/docs/install_linux.md
gh auth login
If
ghis not installed or not authenticated,start_ticketwill return clear instructions instead of silently failing.
Setup
1. Get your Jira API token
- Go to https://id.atlassian.com/manage-profile/security/api-tokens
- Click Create API token
- Copy the token — you'll need it below
2. Configure your MCP client
✨ The easiest way (Web Setup UI):
Simply add the server to your MCP config without any environment variables. When the MCP server starts, it will automatically launch an interactive web UI in your browser (http://localhost:9898). You can enter your Jira credentials there, and it will securely save them directly into your AI IDE's local configuration file and seamlessly start the MCP server!
Supported auto-save clients: VS Code, Cursor, Claude Desktop, and Antigravity.
Or, configure it manually:
Pick your AI IDE below and add the config. Replace the env values with your own.
GitHub Copilot (VS Code)
Open User Settings (JSON) via Cmd+Shift+P → Open User Settings (JSON) and add:
"mcp": {
"servers": {
"jira-mcp": {
"command": "npx",
"args": ["-y", "@abdelrahmanhsn/jira-mcp"],
"env": {
"JIRA_EMAIL": "you@company.com",
"JIRA_TOKEN": "your-api-token",
"JIRA_DOMAIN": "yourcompany.atlassian.net",
"JIRA_PROJECT": "PROJ"
}
}
}
}
Cursor
Open ~/.cursor/mcp.json (or Cursor Settings → MCP) and add:
{
"mcpServers": {
"jira-mcp": {
"command": "npx",
"args": ["-y", "@abdelrahmanhsn/jira-mcp"],
"env": {
"JIRA_EMAIL": "you@company.com",
"JIRA_TOKEN": "your-api-token",
"JIRA_DOMAIN": "yourcompany.atlassian.net",
"JIRA_PROJECT": "PROJ"
}
}
}
}
Claude Desktop
Open ~/Library/Application Support/Claude/claude_desktop_config.json and add:
{
"mcpServers": {
"jira-mcp": {
"command": "npx",
"args": ["-y", "@abdelrahmanhsn/jira-mcp"],
"env": {
"JIRA_EMAIL": "you@company.com",
"JIRA_TOKEN": "your-api-token",
"JIRA_DOMAIN": "yourcompany.atlassian.net",
"JIRA_PROJECT": "PROJ"
}
}
}
}
Environment Variables
| Variable | Required | Description |
|---|---|---|
JIRA_EMAIL |
✅ | Your Jira account email |
JIRA_TOKEN |
✅ | Your Jira API token |
JIRA_DOMAIN |
✅ | Your Jira domain, e.g. yourcompany.atlassian.net |
JIRA_PROJECT |
✅ | Your Jira project key, e.g. PROJ |
Usage Examples
Everyday queries
- "Show me my current Jira tickets"
- "What's in my active sprint?"
- "Get me the details for PROJ-1234"
- "Add a comment to PROJ-123 saying the fix is deployed to staging"
- "Give me my standup for today"
- "Summarize the active sprint — how many tickets are done vs in progress?"
- "Search for open bugs related to login"
PR workflow
- "Get PR context for branch STUD-17891-add-email-icon" → Extracts the ticket key from the branch, fetches description + comments, returns a formatted PR description block ready to paste or expand.
Autonomous mode — start_ticket
The most powerful tool. One prompt and AI does everything:
"Start working on STUD-17931"
What happens automatically, with no stops:
- ✅ Self-assigns the Jira ticket to you
- ✅ Moves it to In Progress
- ✅ Creates and switches to a git branch (e.g.
stud-17931-content-preview-bug) - ✅ AI reads description, acceptance criteria, and comments
- ✅ Implements the feature/fix
- ✅ Runs the test suite — fixes failures automatically
- ✅ Commits and pushes the branch
- ✅ Opens a PR via
gh pr create - ✅ Posts the PR link as a comment on the Jira ticket
You can optionally pass the repo path:
"Start working on STUD-17931 in /Users/you/code/my-project"
If omitted, the tool auto-detects the git repo from the current working directory.
Security
- Credentials are never stored in code — they are injected at runtime by your MCP client
- Each user provides their own credentials in their local MCP config
- Your API token is only sent to your own Jira domain over HTTPS
License
ISC
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。