OpManager MCP Server
Enables AI assistants to interact with ManageEngine OpManager for network monitoring and IT operations management. Provides credential-less access to 85+ API endpoints for managing devices, alarms, discovery, dashboards, and reports through natural language.
README
OpManager MCP Server
A credential-less Model Context Protocol (MCP) server for ManageEngine OpManager REST API integration. This server enables AI assistants like Claude to interact with your OpManager infrastructure through natural language.
✨ Key Features
- 🔐 Credential-less Design: No hardcoded API keys - users provide
hostandapiKeyper request - 🔄 SSL Auto-Detection: Port 8061 → HTTPS, Port 8060 → HTTP (with manual override)
- 📡 85+ API Endpoints: Full OpManager API coverage for devices, alarms, dashboards, discovery, and more
- 🛠 Dynamic Tool Generation: Automatically generates MCP tools from OpenAPI specification
- 🌐 Multiple Transports: Supports stdio (Claude Desktop) and HTTP/SSE (n8n, web clients)
- 🐳 Docker Ready: Containerized deployment with Docker and Docker Compose
🚀 Quick Start
Installation
# Clone the repository
git clone https://github.com/sachdev27/opmanager-mcp-server.git
cd opmanager-mcp-server
# Create virtual environment
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
# Install the package
pip install -e ".[http]"
Start the HTTP Server
uvicorn opmanager_mcp.http_server:app --host 0.0.0.0 --port 3000
Test a Tool Call
curl -X POST http://localhost:3000/call \
-H "Content-Type: application/json" \
-d '{
"name": "opmanager_get_allDevices",
"arguments": {
"host": "opmanager.example.com",
"apiKey": "your-api-key-here",
"port": 8061
}
}'
📋 Configuration
Environment Variables
Create a .env file (optional - for server defaults only):
cp .env.example .env
| Variable | Description | Default |
|---|---|---|
MCP_SERVER_LOG_LEVEL |
Logging level | INFO |
ALLOWED_HTTP_METHODS |
Allowed HTTP methods for tools | GET,POST,PUT,DELETE,PATCH |
LOCAL_OPENAPI_SPEC_PATH |
Path to OpenAPI spec | bundled openapi.json |
Note:
OPMANAGER_HOSTandOPMANAGER_API_KEYare NOT configured server-side. Users provide these per-request for security.
Getting Your OpManager API Key
- Log in to OpManager web console
- Navigate to Settings → REST API
- Generate a new API key
- Use this key in your tool calls
🔧 Tool Parameters
Every tool accepts these connection parameters:
| Parameter | Required | Description |
|---|---|---|
host |
✅ Yes | OpManager server hostname |
apiKey |
✅ Yes | API key for authentication |
port |
No | Server port (default: 8060) |
use_ssl |
No | Force SSL (auto-detected from port) |
verify_ssl |
No | Verify SSL certificates (default: true) |
SSL Auto-Detection
- Port 8061: Automatically uses HTTPS
- Port 8060: Automatically uses HTTP
- Override with
use_ssl: true/falseif needed
🌐 HTTP API Endpoints
| Endpoint | Method | Description |
|---|---|---|
/health |
GET | Health check with tool count |
/tools |
GET | List all available tools |
/sse |
GET | SSE connection for MCP |
/messages |
POST | MCP message handler |
/call |
POST | Direct tool invocation |
Health Check
curl http://localhost:3000/health
# {"status":"healthy","tool_count":60}
List Tools
curl http://localhost:3000/tools | jq '.tools[].name'
🤖 n8n Integration
- Start the HTTP server on port 3000
- In n8n, add an AI Agent node with MCP Client tool
- Configure the MCP Client:
- SSE URL:
http://localhost:3000/sse - Messages URL:
http://localhost:3000/messages
- SSE URL:
Example System Prompt for n8n
You are an IT operations assistant with access to OpManager for network monitoring.
When using OpManager tools, always include:
- host: "opmanager.company.com"
- apiKey: "your-api-key"
- port: 8061 (for HTTPS)
Available operations:
- List all devices: opmanager_get_allDevices
- Get device details: opmanager_get_device (requires deviceName)
- List alarms: opmanager_get_alarms
- Acknowledge alarm: opmanager_add_alarmNotes
🖥 Claude Desktop Integration
Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json):
{
"mcpServers": {
"opmanager": {
"command": "python",
"args": ["-m", "opmanager_mcp.main"],
"cwd": "/path/to/opmanager-mcp-server",
"env": {
"LOCAL_OPENAPI_SPEC_PATH": "/path/to/opmanager-mcp-server/openapi.json"
}
}
}
}
Note: With Claude Desktop, you'll tell Claude your OpManager host and API key in conversation, and it will include them in tool calls.
🛠 Available Tools (60+ GET operations)
Devices
opmanager_get_allDevices- List all monitored devicesopmanager_get_device- Get device details by nameopmanager_get_deviceAvailability- Device availability history
Alarms
opmanager_get_alarms- List alarms with filteringopmanager_get_alarmDetails- Get alarm detailsopmanager_add_alarmNotes- Add notes/acknowledge alarm
Discovery
opmanager_get_discoveryStatus- Check discovery progressopmanager_add_discovery- Start network discovery
Reports & Dashboards
opmanager_get_allDashboards- List all dashboardsopmanager_get_scheduledReports- List scheduled reports
And more...
Run curl http://localhost:3000/tools to see all available tools.
🐳 Docker
Build and Run
docker build -t opmanager-mcp-server .
docker run -d -p 3000:3000 --name opmanager-mcp opmanager-mcp-server
Docker Compose
docker-compose up -d
🧪 Development
Run Tests
# Install dev dependencies
pip install -e ".[dev]"
# Run all tests
pytest
# Run with coverage
pytest --cov=opmanager_mcp --cov-report=term-missing
# Current: 32 tests, 50% coverage
Code Quality
# Format
black opmanager_mcp tests
isort opmanager_mcp tests
# Lint
ruff check opmanager_mcp tests
# Type check
mypy opmanager_mcp
Regenerate OpenAPI Spec
python generate_openapi.py
📁 Project Structure
opmanager-mcp-server/
├── opmanager_mcp/
│ ├── __init__.py # Package exports
│ ├── api_client.py # HTTP client for OpManager API
│ ├── config.py # Configuration management
│ ├── exceptions.py # Custom exceptions
│ ├── http_server.py # HTTP/SSE server (Pure ASGI)
│ ├── logging_config.py # Logging configuration
│ ├── main.py # CLI entry point
│ ├── server.py # MCP server implementation
│ └── tool_generator.py # OpenAPI to MCP tool converter
├── tests/
│ ├── conftest.py # Test fixtures
│ ├── test_api_client.py # API client tests
│ ├── test_config.py # Config tests
│ ├── test_http_server.py # HTTP server tests
│ ├── test_server.py # MCP server tests
│ └── test_tool_generator.py # Tool generation tests
├── openapi.json # OpManager OpenAPI specification
├── pyproject.toml # Project configuration
├── Dockerfile # Container image
├── docker-compose.yml # Compose configuration
└── README.md # This file
📄 License
MIT License - see LICENSE for details.
🙏 Acknowledgments
- Built with Model Context Protocol SDK
- OpManager REST API by ManageEngine
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。