JIRA MCP Server

A Model Context Protocol (MCP) server implementation that provides access to JIRA data with relationship tracking, optimized data payloads, and data cleaning for AI context windows.

ℹ️ There is a separate MCP server for Confluence

Jira Cloud & Jira Server (Data Center) Support

This MCP server supports both Jira Cloud and Jira Server (Data Center) instances. You can select which type to use by setting the JIRA_TYPE environment variable:

• cloud (default): For Jira Cloud (Atlassian-hosted)
• server: For Jira Server/Data Center (self-hosted)

The server will automatically use the correct API version and authentication method for the selected type.

Features

• Search JIRA issues using JQL (maximum 50 results per request)
• Retrieve epic children with comment history and optimized payloads (maximum 100 issues per request)
• Get detailed issue information including comments and related issues
• Create, update, and manage JIRA issues
• Add comments to issues
• Extract issue mentions from Atlassian Document Format
• Track issue relationships (mentions, links, parent/child, epics)
• Clean and transform rich JIRA content for AI context efficiency
• Support for file attachments with secure multipart upload handling
• Supports both Jira Cloud and Jira Server (Data Center) APIs

Prerequisites

• Bun (v1.0.0 or higher)
• JIRA account with API access

Environment Variables

JIRA_API_TOKEN=your_api_token
JIRA_BASE_URL=your_jira_instance_url  # e.g., https://your-domain.atlassian.net
JIRA_USER_EMAIL=your_email
JIRA_TYPE=cloud   # or 'server' for Jira Server/Data Center (optional, defaults to 'cloud')

Installation & Setup

1. Clone the repository:

git clone [repository-url]
cd jira-mcp

2. Install dependencies and build:

bun install
bun run build

3. Configure the MCP server:

Edit the appropriate configuration file:

macOS:

• Cline: ~/Library/Application Support/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json
• Claude Desktop: ~/Library/Application Support/Claude/claude_desktop_config.json

Windows:

• Cline: %APPDATA%\Code\User\globalStorage\saoudrizwan.claude-dev\settings\cline_mcp_settings.json
• Claude Desktop: %APPDATA%\Claude Desktop\claude_desktop_config.json

Linux:

• Cline: ~/.config/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json
• Claude Desktop: sadly doesn't exist yet

Add the following configuration under the mcpServers object:

{
  "mcpServers": {
    "jira": {
      "command": "node",
      "args": ["/absolute/path/to/jira-mcp/build/index.js"],
      "env": {
        "JIRA_API_TOKEN": "your_api_token",
        "JIRA_BASE_URL": "your_jira_instance_url",
        "JIRA_USER_EMAIL": "your_email",
        "JIRA_TYPE": "cloud"
      }
    }
  }
}

4. Restart the MCP server.

Within Cline's MCP settings, restart the MCP server. Restart Claude Desktop to load the new MCP server.

Development

Run tests:

bun test

Watch mode for development:

bun run dev

To rebuild after changes:

bun run build

Project Structure (Updated)

The server has been restructured to follow best practices with organized tool modules:

src/
├── index.ts              # Main server entry point
├── config.ts             # Configuration management  
├── types/
│   └── index.ts          # TypeScript type definitions
├── utils/
│   └── auth.ts           # Authentication utilities
├── tools/                # MCP tools organized by category
│   ├── issues.ts         # Issue management tools
│   ├── projects.ts       # Project management tools
│   ├── users.ts          # User management tools
│   ├── metadata.ts       # Metadata tools (types, priorities, statuses)
│   ├── boards.ts         # Agile/board management tools
│   └── resources.ts      # Resource management tools
├── services/
│   ├── jira-api.ts       # JIRA Cloud API service
│   └── jira-server-api.ts # JIRA Server API service
└── validation/           # Zod validation schemas
    ├── issues/
    ├── projects/
    └── users/

Available MCP Tools

search_issues

Search JIRA issues using JQL. Returns up to 50 results per request.

Input Schema:

{
  searchString: string; // JQL search string
}

get_epic_children

Get all child issues in an epic including their comments and relationship data. Limited to 100 issues per request.

Input Schema:

{
  epicKey: string; // The key of the epic issue
}

get_issue

Get detailed information about a specific JIRA issue including comments and all relationships.

Input Schema:

{
  issueId: string; // The ID or key of the JIRA issue
}

create_issue

Create a new JIRA issue with specified fields.

Input Schema:

{
  projectKey: string, // The project key where the issue will be created
  issueType: string, // The type of issue (e.g., "Bug", "Story", "Task")
  summary: string, // The issue summary/title
  description?: string, // Optional issue description
  fields?: { // Optional additional fields
    [key: string]: any
  }
}

update_issue

Update fields of an existing JIRA issue.

Input Schema:

{
  issueKey: string, // The key of the issue to update
  fields: { // Fields to update
    [key: string]: any
  }
}

add_attachment

Add a file attachment to a JIRA issue.

Input Schema:

{
  issueKey: string, // The key of the issue
  fileContent: string, // Base64 encoded file content
  filename: string // Name of the file to be attached
}

add_comment

Add a comment to a JIRA issue. Accepts plain text and converts it to the required Atlassian Document Format internally.

Input Schema:

{
  issueIdOrKey: string, // The ID or key of the issue to add the comment to
  body: string // The content of the comment (plain text)
}

Data Cleaning Features

• Extracts text from Atlassian Document Format
• Tracks issue mentions in descriptions and comments
• Maintains formal issue links with relationship types
• Preserves parent/child relationships
• Tracks epic associations
• Includes comment history with author information
• Removes unnecessary metadata from responses
• Recursively processes content nodes for mentions
• Deduplicates issue mentions

Technical Details

• Built with TypeScript in strict mode
• Uses Bun runtime for improved performance
• Vite for optimized builds
• Uses JIRA REST API v3
• Basic authentication with API tokens
• Batched API requests for related data
• Optimized response payloads for AI context windows
• Efficient transformation of complex Atlassian structures
• Robust error handling
• Rate limiting considerations
• Maximum limits:
  • Search results: 50 issues per request
  • Epic children: 100 issues per request
• Support for multipart form data for secure file attachments
• Automatic content type detection and validation

Error Handling

The server implements a comprehensive error handling strategy:

• Network error detection and appropriate messaging
• HTTP status code handling (especially 404 for issues)
• Detailed error messages with status codes
• Error details logging to console
• Input validation for all parameters
• Safe error propagation through MCP protocol
• Specialized handling for common JIRA API errors
• Base64 validation for attachments
• Multipart request failure handling
• Rate limit detection
• Attachment parameter validation

LICENCE

This project is licensed under the MIT License - see the LICENCE file for details.

Gmail MCP Server

A Model Context Protocol (MCP) server implementation that provides access to Gmail functionality using Bearer token authentication.

Features

• Send emails - Send emails via Gmail with support for TO, CC, BCC recipients
• Search emails - Search emails using Gmail's powerful search syntax
• Read emails - Retrieve and read email content by message ID
• Delete emails - Delete emails by message ID
• List labels - Get all available Gmail labels
• Bearer token authentication - Secure authentication using Gmail app passwords

Prerequisites

• Node.js (v18 or higher)
• Gmail account with app password enabled
• MCP client (Claude Desktop, Cline, etc.)

Authentication Setup

This server uses Gmail app passwords for authentication. Here's how to set it up:

1. Go to your Google Account settings (myaccount.google.com)
2. Enable 2-factor authentication if not already enabled
3. Go to Security > App passwords
4. Generate a new app password for 'Mail'
5. Configure your MCP client to send Bearer token as: [email protected]:your_16_character_app_password

Installation

From npm

npm install -g gmail-bearer-mcp-server

From source

git clone [repository-url]
cd email
npm install
npm run build

Configuration

For MCP Clients

Edit your MCP client configuration file:

macOS:

• Cline: ~/Library/Application Support/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json
• Claude Desktop: ~/Library/Application Support/Claude/claude_desktop_config.json

Windows:

• Cline: %APPDATA%\Code\User\globalStorage\saoudrizwan.claude-dev\settings\cline_mcp_settings.json
• Claude Desktop: %APPDATA%\Claude Desktop\claude_desktop_config.json

Linux:

• Cline: ~/.config/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json

Add the following configuration:

{
  "mcpServers": {
    "gmail": {
      "command": "gmail-bearer-mcp",
      "env": {
        "GMAIL_BEARER_TOKEN": "[email protected]:your_app_password"
      }
    }
  }
}

Alternatively, you can provide the bearer token via headers:

• Authorization: Bearer [email protected]:your_app_password
• X-Gmail-Token: [email protected]:your_app_password

Available Tools

send_email

Send an email via Gmail.

Parameters:

• to (required): Array of recipient email addresses
• subject (required): Email subject line
• body (required): Email body content (plain text)
• cc (optional): Array of CC email addresses
• bcc (optional): Array of BCC email addresses

Example:

{
  "to": ["[email protected]"],
  "subject": "Test Email",
  "body": "This is a test email from the Gmail MCP server.",
  "cc": ["[email protected]"],
  "bcc": ["[email protected]"]
}

search_emails

Search emails using Gmail's search syntax.

Parameters:

• query (required): Gmail search query
• maxResults (optional): Maximum number of results (1-500, default: 10)

Example queries:

• from:[email protected] - Emails from specific sender
• has:attachment - Emails with attachments
• after:2024/01/01 - Emails after specific date
• subject:important - Emails with "important" in subject

read_email

Read the content of a specific email.

Parameters:

• messageId (required): Gmail message ID

delete_email

Delete an email by message ID.

Parameters:

• messageId (required): Gmail message ID to delete

list_labels

List all available Gmail labels.

Parameters: None

Development

Build

npm run build

Development mode

npm run dev

Start

npm start

Security

• This server uses Gmail app passwords for authentication
• Bearer tokens are extracted from request headers
• No credentials are stored in the server
• All communication uses HTTPS via Gmail's API

Troubleshooting

Authentication Issues

1. "No bearer token available" - Make sure you're providing the bearer token in one of these ways:

   • Authorization: Bearer your_email:your_app_password header
   • X-Gmail-Token: your_email:your_app_password header
   • GMAIL_BEARER_TOKEN=your_email:your_app_password environment variable

2. "Gmail API request failed: 401" - Your app password may be incorrect or expired. Generate a new one.

3. "Gmail API request failed: 403" - Your Gmail account may have restrictions. Check your Google Account security settings.

Common Issues

• Make sure 2-factor authentication is enabled on your Google account
• App passwords only work with 2FA enabled
• Use the full email address followed by colon and the 16-character app password
• Remove any spaces from the app password

License

MIT