Jira MCP Server
Jira MCP Server
一个 MCP(Model Context Protocol)服务器,可将 VS Code Copilot 连接到你的 Jira 实例(Cloud、Server 或 Data Center)。
快速设置
1. 安装并构建
cd jira-mcp-server
npm install
npm run build2. 配置凭据
复制示例文件并填入你的值:
cp .env.example .env编辑 .env:
# Your Jira site URL
JIRA_BASE_URL=https://yourcompany.atlassian.net
# Your email (Cloud) or username (Server)
JIRA_EMAIL=you@yourcompany.com
# API token (Cloud) or Personal Access Token (Server)
JIRA_API_TOKEN=your-token-here
# "cloud" or "server"
JIRA_AUTH_TYPE=cloud如何获取你的令牌
平台 | 获取方式 |
Atlassian Cloud | 前往 API 令牌 → 创建 API 令牌 |
Jira Server/DC | 前往你的 Jira 个人资料 → 个人访问令牌 → 创建令牌 |
3. 连接到 VS Code
将此添加到你的 VS Code settings.json(或工作区 .vscode/settings.json):
{
"mcp": {
"servers": {
"jira": {
"command": "node",
"args": ["/absolute/path/to/jira-mcp-server/dist/index.js"],
"env": {
"JIRA_BASE_URL": "https://yourcompany.atlassian.net",
"JIRA_EMAIL": "you@yourcompany.com",
"JIRA_API_TOKEN": "your-token-here",
"JIRA_AUTH_TYPE": "cloud"
}
}
}
}
}提示: 将
/absolute/path/to/jira-mcp-server替换为你机器上的实际路径。
Related MCP server: Jira Stories MCP Server
可用工具
工具 | 描述 |
| 使用 JQL 搜索问题 |
| 按 key 获取问题的完整详情 |
| 创建新问题 |
| 更新现有问题的字段 |
| 向问题添加评论 |
| 列出可用的状态转换 |
| 将问题移动到新状态 |
| 将问题分配给用户 |
| 列出所有可访问的项目 |
| 获取已认证用户的个人资料 |
示例提示(在 Copilot Chat 中)
"显示分配给我的 Jira 问题"
"获取 PROJ-123 的详情"
"在项目 PROJ 中创建缺陷:登录页面在 Safari 上崩溃"
"将 PROJ-456 移至进行中"
"向 PROJ-789 添加评论:已在提交 abc123 中修复"
Available Tools
10 toolsadd_commentC
Add a comment to a Jira issue.
Inputs: body (required), issueKey (required).
assign_issueA
Assign a Jira issue to a user. For Cloud use accountId, for Server use username.
Inputs: assignee (required), issueKey (required).
create_issueB
Create a new Jira issue in a project.
Inputs: summary (required), projectKey (required), issueType, description.
get_issueA
Get full details of a single Jira issue by its key (e.g. PROJ-123).
Inputs: issueKey (required).
get_myselfA
Get the profile of the currently authenticated Jira user.
Inputs: no named parameters.
get_transitionsA
Get available status transitions for an issue (useful before transitioning).
Inputs: issueKey (required).
list_projectsA
List all Jira projects accessible to the authenticated user.
Inputs: no named parameters.
search_issuesA
Search Jira issues using JQL. Returns key, summary, status, assignee, priority, and type.
Inputs: jql (required), maxResults.
transition_issueA
Transition a Jira issue to a new status. Use get_transitions first to find the right transition ID.
Inputs: issueKey (required), transitionId (required), comment.
update_issueB
Update fields on an existing Jira issue.
Inputs: fields (required), issueKey (required).
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
10 tool updates
v1.0.0- First observed
add_comment - First observed
assign_issue - First observed
create_issue - First observed
get_issue - First observed
get_myself - First observed
get_transitions - First observed
list_projects - First observed
search_issues - First observed
transition_issue - First observed
update_issue
TDQS
Scored across 10 tools
Each tool has a clear, distinct purpose: search vs. get, create vs. update, comment, transitions, assignment, project listing, and user profile. There is no ambiguity between tools; an agent can reliably select the correct one based on the action needed.
All tools follow a consistent verb_noun pattern in snake_case (e.g., search_issues, get_issue, create_issue, add_comment). Even the 'get_*' tools differ by object (issue, transitions, myself), maintaining predictability. No mixed conventions or vague verbs.
With 10 tools, the server is well-scoped for Jira's core functionality. Each tool serves a necessary purpose, covering create, read, update, search, comments, transitions, assignment, and user/project context. No redundancy or bloat.
The set covers the primary Jira workflows: issue lifecycle (create, get, update, transition, assign), search, comments, and context (projects, user). Minor gaps exist (e.g., no delete_issue, no link_issue), but these are not critical for most automation tasks and can be worked around.
Maintenance
Related MCP Connectors
Connect to Atlassian Jira, Confluence, Loom, and more to search, create, and manage your work.
Connect AI assistants to GitHub - manage repos, issues, PRs, and workflows through natural language.
Run accessibility, contrast, and design token audits and file QA issues from your AI coding tool.
Operate GitHub from your AI assistant: repos, PRs, issues, Actions. Previews and read-only mode.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceEnables AI assistants to search, view, create, and update JIRA issues using natural language commands and JQL queries.1,383 npmApache 2.0
- AlicenseNot gradedqualityDmaintenanceConnects Jira projects to AI assistants, enabling natural language queries to view, search, and filter stories with real-time access to descriptions, comments, status, and metadata without leaving your development environment.1,383 npmApache 2.0
- AlicenseNot gradedqualityDmaintenanceEnables searching, reading, listing, creating, and updating Confluence pages via natural language in GitHub Copilot.8 npmMIT
- AlicenseNot gradedqualityCmaintenanceSupercharge your AI assistant with seamless Jira integration. Enables AI assistants like GitHub Copilot and Claude to interact with your Jira Cloud instance.874 npmMIT