mcp-ninjaone
An MCP server that connects AI assistants to the NinjaOne remote monitoring and management platform via the REST API v2. It provides tools for device inventory, organization management, alert handling, maintenance scheduling, and automated job execution.
README
mcp-ninjaone
A Model Context Protocol (MCP) server for NinjaOne (formerly NinjaRMM), providing AI assistants with full access to remote monitoring and management through the NinjaOne REST API v2.
Features
19 tools across five categories:
Device Management
| Tool | Description |
|---|---|
list_devices |
List all managed devices with online status, OS, organization, and last contact |
get_device |
Get detailed device info including hardware, agent version, and network details |
get_device_activities |
Get recent activity log for a device (alerts, scripts, status changes) |
get_device_os_patch_status |
View pending, installed, and failed OS patches for a device |
get_device_software |
List all installed software on a device |
get_device_processor_info |
Get CPU/processor details for a device |
get_device_disks |
Get disk/storage info including capacity and free space |
Organizations
| Tool | Description |
|---|---|
list_organizations |
List all organizations (clients/customers) with ID, name, and node count |
get_organization |
Get detailed organization info including contacts, tags, and policy assignments |
list_organization_devices |
List all devices belonging to a specific organization |
get_organization_locations |
Get configured locations for an organization |
get_organization_policies |
Get policy assignments for an organization |
Alerts & Conditions
| Tool | Description |
|---|---|
list_alerts |
List active alerts across all devices, filterable by severity |
list_device_alerts |
List all active alerts for a specific device |
reset_alert |
Reset (acknowledge and clear) an active alert by its UID |
list_alert_conditions |
List all configured alert trigger conditions |
Maintenance Windows
| Tool | Description |
|---|---|
list_maintenance_windows |
List all scheduled maintenance windows |
create_maintenance_window |
Create a maintenance window for a device to suppress alerts |
cancel_device_maintenance |
Cancel an active maintenance window to re-enable alerting |
Jobs & Automation
| Tool | Description |
|---|---|
list_running_jobs |
List currently running jobs across all devices |
get_device_jobs |
Get all jobs (running, completed, failed) for a specific device |
run_script_on_device |
Execute a saved script on a device |
Installation
git clone git@github.com:fredriksknese/mcp-ninjaone.git
cd mcp-ninjaone
npm install
npm run build
Configuration
The server is configured via environment variables:
| Variable | Required | Description |
|---|---|---|
NINJAONE_CLIENT_ID |
Yes | OAuth2 client ID from NinjaOne API credentials |
NINJAONE_CLIENT_SECRET |
Yes | OAuth2 client secret from NinjaOne API credentials |
NINJAONE_INSTANCE |
Yes | NinjaOne instance hostname (see table below) |
Instance Hostnames
| Region | Hostname |
|---|---|
| US | app.ninjarmm.com |
| EU | eu.ninjarmm.com |
| Oceania | oc.ninjarmm.com |
| Canada | ca.ninjarmm.com |
Creating API Credentials
- Log in to your NinjaOne instance
- Navigate to Administration > Apps > API
- Click Add to create a new API application
- Select Client Credentials as the authorization flow
- Grant the required scopes:
monitoring,management,control,offline_access - Copy the Client ID and Client Secret
Usage with Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"ninjaone": {
"command": "node",
"args": ["/absolute/path/to/mcp-ninjaone/dist/index.js"],
"env": {
"NINJAONE_CLIENT_ID": "your-client-id",
"NINJAONE_CLIENT_SECRET": "your-client-secret",
"NINJAONE_INSTANCE": "app.ninjarmm.com"
}
}
}
}
Usage with Claude Code
Add to your Claude Code MCP settings:
claude mcp add ninjaone -- node /absolute/path/to/mcp-ninjaone/dist/index.js
Set the required environment variables before running, or configure them in your MCP settings.
Example Prompts
Once connected, you can ask your AI assistant things like:
- "Show me all devices that are currently offline"
- "What active alerts do we have across all organizations?"
- "Put device 1234 into maintenance mode for the next 2 hours"
- "List all devices for the Acme Corp organization"
- "What software is installed on device 5678?"
- "Show me recent activity for device 1234"
- "Run the cleanup script on device 9012"
- "What jobs are currently running across all devices?"
- "Reset all minor alerts for device 1234"
- "What patches are pending on device 5678?"
Development
npm run dev # Run with tsx (auto-reloads)
npm run build # Compile TypeScript to dist/
npm start # Run compiled output
Architecture
src/
├── index.ts # Entry point — creates MCP server + STDIO transport
├── ninjaone-client.ts # HTTP client with OAuth2 client credentials flow
└── tools/
├── devices.ts # Device management and inventory (7 tools)
├── organizations.ts # Organization/client management (5 tools)
├── alerts.ts # Alert monitoring and acknowledgement (4 tools)
├── maintenance.ts # Maintenance window management (3 tools)
└── jobs.ts # Job execution and monitoring (3 tools)
The client handles OAuth2 token lifecycle automatically — tokens are fetched on first use and refreshed when they expire (typically after 1 hour).
Requirements
- Node.js 18+
- NinjaOne account with API access enabled
License
SEE LICENSE IN LICENSE
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。