Jira MCP Server
以下摘录自该仓库的 README(片段) · 在 GitHub 读完整 README →
The most complete **open source** Model Context Protocol (MCP) server for **Jira & Atlassian**. Connect any MCP-compatible AI client to your Jira instance in seconds — manage issues, sprints, comments, transitions, users, and Confluence pages without leaving your AI assistant.
Compatible AI Clients
| Client | Supported |
|--------|-----------|
| Claude Desktop | ✅ |
| Claude Code | ✅ |
| Cursor | ✅ |
| Windsurf | ✅ |
| Cline | ✅ |
| Continue | ✅ |
| Any MCP-compatible client | ✅ |
Find This Server On
- MCP Market — MCP server marketplace
- MCP Servers — Awesome MCP Servers directory
- Claude Marketplaces — Claude plugins & MCP directory
- npm — npx raalarcon-jira-mcp-server
Features
- ✅ **Project Management**: List projects and issue types
- ✅ **Issue CRUD**: Create, read, update and delete issues
- ✅ **Comments**: Create, read, update and delete comments with enhanced pagination
- ✅ **Transitions**: Move issues between states
- ✅ **Assignments**: Assign issues to users
- ✅ **User Management**: Search and manage users
- ✅ **Sprint Management**: Complete agile sprint lifecycle management
- ✅ **Wiki Integration**: Access Confluence pages by URL identifier with HTML to text conversion
- ✅ **Rich Text Support**: Markdown to ADF conversion for formatted descriptions and comments
- ✅ **Validation**: Yup schema validation
- ✅ **Authentication**: Full Jira Cloud support
- ✅ **Optimized Responses**: Token-efficient field filtering
- ✅ **Type Safety**: Full TypeScript support
Installation
Option 1: Using npx (Recommended)
The easiest way to use this MCP server is with npx:
1. **Configure your MCP client** (e.g., Claude Desktop) with this configuration:
2. **Get your Jira API token** (see instructions below)
That's it! The server will be automatically downloaded and run when needed.
Option 2: Local Development
1. **Clone the repository**:
2. **Install dependencies**:
3. **Build the project**:
4. **Configure your MCP client** with the full path to the built server:
**Note**: Replace /full/path/to/jira-mcp/ with the actual absolute path to your project directory.
Getting API Token
1. Go to Atlassian Account Settings
2. Click "Create API token"
3. Give it a descriptive name
4. Copy the generated token
Usage
Using with npx (Recommended)
Once configured in your MCP client, the server will automatically start when needed. No additional setup required!
Rich Text Support with Markdown
The server now supports **automatic Markdown to ADF conversion** for issue descriptions and comments. Simply use Markdown syntax and it will be automatically converted to Atlassian Document Format (ADF).
Supported Markdown Elements
- **Headers**: # H1, ## H2, ### H3
- **Text formatting**: **bold**, *italic*
- **Code**: inline code and code blocks
- **Lists**: - bullet lists and 1. numbered lists
- **Links**: text
- **Blockquotes**: > quoted text
- **Checkboxes**: - [x] completed task
Example Usage
Development
Production
Testing with npx
You can also test the server directly with npx:
Available Tools
Projects
get_projects
Retrieves all projects accessible to the authenticated user.
**Parameters**:
- expand (optional): Additional data to include
- recent (optional): Number of recent projects (0-20)
**Response**: Array of projects with essential fields:
get_issue_types
Gets all available issue types for a specific project.
**Parameters**:
- projectKey (required): Project key
**Response**: Array of issue types with essential fields:
Issues
create_issue
Creates a new issue in Jira.
**Parameters**:
- projectKey (required): Project key
- summary (required): Issue summary
- issueType (required): Issue type (Bug, Story, Task, etc.)
- description (optional): Issue description
- priority (optional): Priority (Highest, High, Medium, Low, Lowest)
- assignee (optional): Assignee account ID
- labels (optional): Array of labels
- components (optional): Array of components
- fixVersions (optional): Array of fix versions
- customFields (optional): Custom field values
**Response**: Issue PROJ-123 created successfully
get_issue
Gets details of a specific issue (custom fields removed for token efficiency).
**Parameters**:
- issueKey (required): Issue key (e.g., PROJ-123)
- expand (optional): Additional information
- fields (optional): Specific fields to return