ops-mcp-server

ops-mcp-server: an AI-driven IT operations platform that fuses LLMs and MCP architecture to enable intelligent monitoring, anomaly detection, and natural human-infrastructure interaction with enterprise-grade security and scalability.

• Project Overview
• Key Features
• Demo Videos
• Installation
• Deployment
• Local MCP Server Configuration
• Interactive Client Usage
• License
• Notes

ops-mcp-server is an IT operations management solution for the AI era. It achieves intelligent IT operations through the seamless integration of the Model Context Protocol (MCP) and Large Language Models (LLMs). By leveraging the power of LLMs and MCP's distributed architecture, it transforms traditional IT operations into an AI-driven experience, enabling automated server monitoring, intelligent anomaly detection, and context-aware troubleshooting. The system acts as a bridge between human operators and complex IT infrastructure, providing natural language interaction for tasks ranging from routine maintenance to complex problem diagnosis, while maintaining enterprise-grade security and scalability.

• Real-time CPU, memory, disk inspections.
• System load and process monitoring.
• Service and network interface checks.
• Log analysis and configuration backup.
• Security vulnerability scans (SSH login, firewall status).
• Detailed OS information retrieval.

• Container, image, and volume management.
• Container resource usage monitoring.
• Log retrieval and health checks.

• Multi-vendor support (Cisco, Huawei, H3C).
• Switch port, VLAN, and router route checks.
• ACL security configuration analysis.
• Optical module and device performance monitoring.

• Extensible plugin architecture.
• Batch operations across multiple devices.
• Tool listing and descriptive commands.

On Cherry Studio

On Terminal

Ensure you have Python 3.10+ installed. This project uses uv for dependency and environment management.

curl -LsSf https://astral.sh/uv/install.sh | sh

uv venv .venv

# Activate the environment
source .venv/bin/activate      # Linux/macOS
.\.venv\Scripts\activate       # Windows

uv pip install -r requirements.txt

Dependencies are managed via pyproject.toml.

cd server_monitor_sse

# Install dependencies
pip install -r requirements.txt

# Start service
cd ..
uv run server_monitor_sse --transport sse --port 8000

Ensure Docker and Docker Compose are installed.

cd server_monitor_sse
docker compose up -d

# Check status
docker compose ps

# Logs monitoring
docker compose logs -f

Add this configuration to your MCP settings:

{
  "ops-mcp-server": {
    "command": "uv",
    "args": [
      "--directory", "YOUR_PROJECT_PATH_HERE",
      "run", "server_monitor.py"
    ],
    "env": {},
    "disabled": true,
    "autoApprove": ["list_available_tools"]
  },
  "network_tools": {
    "command": "uv",
    "args": [
      "--directory", "YOUR_PROJECT_PATH_HERE",
      "run", "network_tools.py"
    ],
    "env": {},
    "disabled": false,
    "autoApprove": []
  },
}

Note: Replace YOUR_PROJECT_PATH_HERE with your project's actual path.

An interactive client (client.py) allows you to interact with MCP services using natural language.

uv pip install openai rich

Edit these configurations within client.py:

# Initialize OpenAI client
self.client = AsyncOpenAI(
    base_url="https://your-api-endpoint",
    api_key="YOUR_API_KEY"
)

# Set model
self.model = "your-preferred-model"

uv run client.py [path/to/server.py]

Example:

uv run client.py ./server_monitor.py

• help - Display help.
• quit - Exit client.
• clear - Clear conversation history.
• model <name> - Switch models.

This project is licensed under the MIT License.

• Ensure remote SSH access is properly configured.
• Adjust tool parameters based on actual deployment conditions.
• This project is under active development; feedback and contributions are welcome.