jira-service-desk-mcp

jira-service-desk-mcp

Connects AI assistants to Jira Service Management instances, providing 40 tools for managing service desks, requests, comments, customers, organizations, queues, SLA, approvals, and attachments.

Category
访问服务器

README

jira-service-desk-mcp

PyPI Python License: MIT

MCP (Model Context Protocol) server for Jira Service Management (formerly Jira Service Desk).
Connects AI assistants — Claude Code, Claude Desktop, Cursor, and others — to your JSM instance via Anthropic's MCP.

Built on top of atlassian-python-api and inspired by mcp-atlassian.

Features

Category Tools
Service Desks List desks, get desk by ID, app info
Request Types List types, get type details, get required fields
Customer Requests Create, get, list my requests, get status
Comments Add public/internal comments, list, get by ID
Customers Create customer, list, add/remove from desk
Participants List, add, remove request participants
Transitions List available transitions, perform transition
Organizations CRUD organizations, manage users, link to desks
Queues List queues, list issues in queue
SLA Get SLA info per request, by metric ID
Approvals List approvals, get by ID, approve/decline
Attachments Upload temp file, attach to request

40 tools total. Write operations are gated by READ_ONLY_MODE.

Quick Start

Install from PyPI

pip install jira-service-desk-mcp

Or with uv:

uv pip install jira-service-desk-mcp

Or install from source:

pip install -e .

Configure

Copy .env.example to .env and fill in your credentials:

cp .env.example .env

Recommended shared format (same env shape as mcp-atlassian):

JIRA_URL=https://jira.your-company.com
JIRA_PERSONAL_TOKEN=<base64-encoded username:token>

Use a base64-encoded username:token string when you want one MCP config style that works across both mcp-atlassian and jira-service-desk-mcp for Atlassian Cloud. On non-Cloud/self-hosted Jira instances, JIRA_PERSONAL_TOKEN now defaults to Bearer/PAT handling to avoid misclassifying base64-looking PATs. If you intentionally need basic auth there, set JIRA_PERSONAL_TOKEN_MODE=basic.

Alternative: Atlassian Cloud (username + API token):

JIRA_URL=https://your-instance.atlassian.net
JIRA_USERNAME=your-email@example.com
JIRA_API_TOKEN=your-api-token

Generate an API token at https://id.atlassian.com/manage-profile/security/api-tokens.

Alternative: Server / Data Center (raw Personal Access Token):

JIRA_URL=https://jira.your-company.com
JIRA_PERSONAL_TOKEN=your-personal-access-token

The token can be either a raw PAT (bearer token), the full Bearer <token> header value, or a base64-encoded username:api_token string. The server strips Bearer automatically when present.

For non-Cloud/self-hosted Jira, JIRA_PERSONAL_TOKEN is treated as Bearer/PAT by default, even if it looks like base64. This avoids accidental fallback to Basic auth for PATs that happen to decode cleanly. If you need to force Basic auth from JIRA_PERSONAL_TOKEN, set:

JIRA_PERSONAL_TOKEN_MODE=basic

Note: JIRA_IS_CLOUD is auto-detected from the URL (.atlassian.net → Cloud). Set it explicitly only if the heuristic is wrong for your instance.

Run

stdio (default — for Claude Code, Claude Desktop, Cursor, etc.):

jira-service-desk-mcp

SSE (for remote/web clients):

jira-service-desk-mcp --transport sse --port 8000

Claude Code Integration

The fastest way to connect JSM to Claude Code:

claude mcp add jira-service-desk -- jira-service-desk-mcp

With environment variables inline:

claude mcp add jira-service-desk \
  -e JIRA_URL=https://jira.your-company.com \
  -e JIRA_PERSONAL_TOKEN=<base64-encoded username:token> \
  -- uvx jira-service-desk-mcp

To scope the server to your project only:

claude mcp add --scope project jira-service-desk \
  -e JIRA_URL=https://jira.your-company.com \
  -e JIRA_PERSONAL_TOKEN=<base64-encoded username:token> \
  -- uvx jira-service-desk-mcp

Verify it works:

claude mcp list

Shared MCP Config

Add the same shape to ~/.claude.json, project-level .mcp.json, claude_desktop_config.json, or .cursor/mcp.json:

{
  "mcpServers": {
    "mcp-atlassian": {
      "type": "stdio",
      "command": "uvx",
      "args": ["mcp-atlassian"],
      "env": {
        "JIRA_URL": "https://jira.your-company.com",
        "JIRA_PERSONAL_TOKEN": "<base64-encoded username:token>"
      }
    },
    "jira-service-desk": {
      "type": "stdio",
      "command": "uvx",
      "args": ["jira-service-desk-mcp"],
      "env": {
        "JIRA_URL": "https://jira.your-company.com",
        "JIRA_PERSONAL_TOKEN": "<base64-encoded username:token>"
      }
    }
  }
}

This mirrors the mcp-atlassian configuration style for Atlassian Cloud while still allowing jira-service-desk-mcp to distinguish Cloud basic-auth tokens from non-Cloud Bearer/PAT tokens.


Environment Variables

Variable Required Description
JIRA_URL Yes Jira instance URL
JIRA_USERNAME Cloud Email for Atlassian Cloud
JIRA_API_TOKEN Cloud API token for Cloud auth
JIRA_PERSONAL_TOKEN Recommended Raw PAT or base64-encoded username:token
JIRA_PERSONAL_TOKEN_MODE No auto (default), pat, or basic override for JIRA_PERSONAL_TOKEN
JIRA_PASSWORD Basic auth password (not recommended)
JIRA_IS_CLOUD No Auto-detected from URL; set explicitly to override
JIRA_SSL_VERIFY No true (default) or false
READ_ONLY_MODE No true to disable all write operations

Available Tools

Service Desks

  • get_service_desk_info — Application info
  • list_service_desks — All accessible service desks
  • get_service_desk — Service desk by ID

Request Types

  • list_request_types — Request types for a desk
  • get_request_type — Specific request type
  • get_request_type_fields — Fields required for a request type

Customer Requests

  • create_customer_request — Create a new request
  • get_customer_request — Get request by key
  • list_my_customer_requests — Your requests
  • get_customer_request_status — Request status

Comments

  • add_request_comment — Add public or internal comment
  • list_request_comments — List comments on a request
  • get_request_comment — Get comment by ID

Customers

  • create_customer — Create customer account
  • list_customers — List customers of a desk
  • add_customers_to_service_desk — Add customers
  • remove_customers_from_service_desk — Remove customers

Participants

  • list_request_participants — List participants
  • add_request_participants — Add participants
  • remove_request_participants — Remove participants

Transitions

  • list_customer_transitions — Available transitions
  • perform_transition — Execute a transition

Organizations

  • list_organizations — List organizations
  • get_organization — Organization by ID
  • create_organization — Create organization
  • delete_organization — Delete organization
  • add_organization_to_service_desk — Link org to desk
  • remove_organization_from_service_desk — Unlink org
  • list_users_in_organization — Users in org
  • add_users_to_organization — Add users to org
  • remove_users_from_organization — Remove users from org

Queues

  • list_queues — Queues in a service desk
  • list_issues_in_queue — Issues in a queue

SLA

  • get_request_sla — SLA info for a request
  • get_request_sla_by_id — Specific SLA metric

Approvals

  • list_approvals — Approvals for a request
  • get_approval — Approval by ID
  • answer_approval — Approve or decline

Attachments

  • upload_temporary_attachment — Upload temp file
  • attach_file_to_request — Attach to request

Architecture

src/mcp_jira_service_desk/
├── __init__.py      # CLI entry point (click)
├── __main__.py      # python -m support
├── config.py        # Environment-based configuration
├── client.py        # ServiceDeskClient — wraps atlassian-python-api
├── formatting.py    # Human-readable formatting for API responses
└── server.py        # FastMCP server with all tool definitions
  • config.py — reads JIRA_* env vars, determines auth type
  • client.py — thin typed wrapper around atlassian.ServiceDesk, normalizes paginated responses
  • formatting.py — converts raw API dicts into markdown for LLM consumption
  • server.pyFastMCP instance with @mcp.tool() decorators for each operation

Development

git clone https://github.com/CzSadykov/jira-service-desk-mcp.git
cd jira-service-desk-mcp
pip install -e ".[dev]"
pytest

License

MIT

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选